','1 ( 0 ISSN 0097-4463 AS. 3(o prx ANNALS of CARNEGIE MUSEUM THE CARNEGIE MUSEUM OF NATURAL HISTORY 4400 FORBES AVENUE • PITTSBURGH, PENNSYLVANIA 15213 VOLUME 70 27 FEBRUARY 2001 NUMBER 1 CONTENTS ARTICLE Late Eocene-Oligocene nonmarine mollusks of the Northern Kishenehn Basin, Montana and British Columbia Harold G. Pierce and Kurt N. Constenius 1 Editors, ANNALS, BULLETIN and SPECIAL PUBLICATIONS: D. S. Berman, Ph.D. D. R. Watters, Ph.D. J. R. Wible, Ph.D. W. O. Goehring, Managing Editor Manuscripts, subscriptions, orders for individual numbers, and changes of address should be sent to: Office of Scientific Publications Carnegie Museum of Natural History 4400 Forbes Avenue Pittsburgh, PA 15213-4080 Phone (412) 622-3287 Fax (412) 622-8837 E-mail: scipubs@clpgh.org ANNALS OF CARNEGIE MUSEUM is published quarterly by The Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania 15213-4080, by the authority of the Board of Trustees of Carnegie Institute. © 2001 Carnegie Institute. @ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 1, Pp. 1-112 27 February 2001 LATE EOCENE^OLIGOCENE NONMARINE MOLLUSKS OF THE NORTHERN KISHENEHN BASIN, MONTANA AND BRITISH COLUMBIA Harold G. Pierce^ Research Associate, Section of Vertebrate Paleontology Kurt N. Constenius^ Research Associate, Section of Vertebrate Paleontology Abstract The Kishenehn Basin is a narrow, asymmetric graben, developed during an episode of crustal extension occurring from middle Eocene to early Miocene time (ca. 49-20 Ma). The Flathead listric normal fault system defines the northeast margin of the basin. The southwest margin is either bounded by antithetic normal faults or is onlapped by Kishenehn Basin strata. High rates of basin subsidence and synextensional sedimentation resulted in an immense thickness of middle Eocene-early Miocene strata (stratigraphic thickness, —5000 m) and a relatively uninterupted depositional record. Provenance and paleocurrent studies indicate that the ancestral Livingston, Lewis, and Clark ranges to the east contributed not only vast amounts of detritus to the Basin, but also ample runoff that included seasonal snow-melt. Sediment ages of outcrops within the basin are linearly banded, with the course of the Flathead River exposing late Eocene-early Oligocene sediments (Chadronian-Orellan NALMA’s), and with the oldest sediments on the southwest and youngest on the east margins of the basin, respectively. An unnamed tephra at the north end of the basin provides a date of 33.0±1.0 Ma (early Oligocene, Orellan NALMA). The molluscan fauna is extremely diverse, 55 taxa (35 terrestrial, 20 aquatic), of which 32 taxa are new, and seven are revised taxonomically. This fauna is clearly divisible into three Groups. Group I is a relict tropical wet fauna, existing along waterways in the sheltered environment of the basin. It may typify the pre-middle Eocene (early Duchesnean NALMA) tropical climate of the Western Interior of North America. Modern analogs of these taxa are found in the Caribbean, Central and South America. Group II is a semitropical, semiarid fauna that developed after the climatic changes of the middle Eocene. Modern analogs are found in a band from the north coast of the Gulf of Mexico across the southern United States and northern Mexico to Southern and Baja California. Essentially 100% of the Group I and II analog taxa are now extra-limital, displaced to the south. Group III taxa developed in uplands along the east margin of the basin. Modern analogs of these taxa are common in the east-central United States. The coexistence of these climatically disparate groups of mollusks is a reflection of the large paleotopographic relief that existed between the basin floor and the eastern mountain peaks, which rose at least 2,000 meters above the basin floor, and permitted microclimates that varied from arid subtropical in the basin to humid temperate in the uplands. At the older localities. Group III taxa were, probably, transported into the basin by run-off from the mountains, but during the climatic cooling of the latest Eocene-early Oligocene, Groups I and II taxa were replaced by Group III taxa in the low-lands. Group III analog taxa are 87% endemic, although their modern centers of distribution are displaced eastward into an area of greater precipitation. The following taxa are described herein as new: Bivalvia: Veneroidea: Sphaeridae- Sphaerium discus. Gastropoda: Archaeogastropoda: Ceresidae- Tozerpina buttsi parva, T. lends-, Helicinidae- Lucidella salishora, L. columbiana-, Mesogastropoda: Hydrobiidae- Tryonia russelli, Cincinatial bow- manana, Fluminicola calderense-, Valvatidae— Valvata procera spadosa-, Basommatophora: Planorbi- dae- Helisoma triangulata, Planorbella fordensis, Biomphalaria spira, Menetus textus, Menetus hilli', Lymnaeidae- Lymnaea bowmanana, L. lacerta; Stylommatophora: Succineidae- Oxylomal kintlana, Omalonyx cocleare; Pupillidae- Pupoides costatus, Gastrocopta kintlana, G. akokala. Vertigo con- steniusi, V. doliara, V. micra-, Valloniidae- Vallonia kootenayorum; Urocoptidae- Holospira tabrumi. *617 Bogey Avenue, O’Neill, Nebraska 68763. ^1622 W. Avenida de Maximillian, Tucson, Arizona 85704. Submitted 30 October 1999. 1 2 Annals of Carnegie Museum VOL. 70 H. heardi, Coelostemma dawsonae; Polygyridae- Praticolella lucifera, Ashmunella(?) sp.; Oreoheli- cidae- Oreohelix dawsonae. Radiocentrum kintlana; Humboldtianidae- Skinnerelix rothi; Helminth- oglyptidae- Xeranionta constenii. The following constitute new taxonomic combinations: Tozerpina “Triodopsis” buttsi (Russell, 1956), Omalonyx “Binneya” antiqua (Russell, 1956), Haplotrema “An- guispira” simplex (Russell, 1956), Lymnaea “Stagnicola” newmarchi (Russell, 1952), Biomphalaria “Planorhis" kishenensis (Russell, 1952), Valvata “Gyraulus” procera (Russell, 1952). Key Words: Eocene-Oligocene, nonmarine mollusks, paleoclimates, structure, crustal extension Introduction General Introduction. — This paper is but a part of a comprehensive, multi- disciplinary study of the Kishenehn Basin, a complex, narrow, half-graben filled by Tertiary sediments, about 150 km in length, trending north to northwest across the border between British Columbia and Montana (Figs. 1, 2). In this paper, we deal with the geology and with the mollusks of only the northern portion of the Kishenehn Basin, that part exposed by the North Fork of the Flathead River in both British Columbia, Canada and Montana, U.S.A. The mollusks of the Middle Fork portion of the Kishenehn Basin, which are exposed by the Middle Fork of the Flathead River and are exclusively in Montana, and the ostracods of the entire basin, will be the subject of future papers (our next report). In this paper, Con- stenius is responsible for the structure and stratigraphy, whereas Pierce is respon- sible for the molluscan paleontology. Recently, there has been renewed interest in the late Eocene-Oligocene-early Miocene nonmarine mollusks of the northern Rocky Mountains. A first report on the Kishenehn Basin by Daly (1912) included a short list of freshwater mollusks, examined by T. W. Stanton, from a locality just north of the International Bound- ary. Stanton felt their age could be as old as Eocene. MacKenzie (1916:36-37) reported observations by W. H. Dali on freshwater mollusks collected at an un- stated location in the Kishenehn Basin. Dali, too, suggested a possible Eocene age for these mollusks. This was followed by the first comprehensive investigation by Russell (1952, 1956, 1964), along the North Fork of the Flathead River, who proposed an age of latest Eocene. Ross (1959:68-72) reported on mollusks iden- tified by T. C. Yen and D. W. Taylor from both the North and Middle Forks of the Flathead River, but with considerable disagreement as to age. While Taylor concurred with an Eocene or Oligocene age. Yen regarded them as “possibly of late Tertiary age” to as young as early Pleistocene (fide Ross, loc. cit.). Since then, several other faunas in western Montana have received detailed appraisal (Table 1, Fig. 3). Roth (1986) studied the Eocene-Oligocene Three Forks molluscan local fauna (l.f.) of Jefferson and Gallatin Counties and the Oligocene Deep River l.f. of Meagher County (Roth and Emberton, 1994). Roth was first to comment extensively on the extralimital nature of these faunas as part of his study of the Three Forks l.f. (loc. cit.). Pierce investigated the late Oligocene- Miocene Cabbage Patch l.f. of Granite and Powell Counties (Pierce and Rasmus- sen, 1992; Pierce, 1992, 1993), and the small Miocene Flint Creek l.f. of Granite County (Pierce and Rasmussen, 1989). Evanoff briefly mentioned the mollusks of the White River l.f., late Eocene-Oligocene, of the Douglas County, Wyoming area (Evanoff et al., 1992). Pierce made preliminary comments on a reconnais- sance of that portion of the Kishenehn Basin exposed by the Middle Fork of the Flathead River (Constenius et al., 1989). Previous investigations of the molluscan faunas of the Kishenehn Basin/Flat- head River, except those of Russell (1952; 1956), had been superficial, or dealt 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 3 Fig. 1. — Landsat image of parts of northwest Montana, southeast British Columbia, and southwest Alberta, showing mountain ranges and other geographic features surrounding the Kishenehn Basin. See Figure 2, upper right, for index map. 4 Annals of Carnegie Museum VOL. 70 CMCl CouUirey u Creek 0NFF31E31 {Chadronian?} ^28 ■> 'nFF4 S WNFF6 1E21 NFF9^ NFFIO' KSHl {Chadronum-i)rel!an ■'/ NFF12 ■•/Out^snean f'hadniman^U i:^KTL3 I ^7 /eariy AnkareeanI mTL2 {Orellm,} KTLl (OrrtUm}/ NFF17- NFF18- INFF19 liNFi^ao C NFF21- NFF23 NFF24 NFF26 NFF28 J^25 Moose BWN2 Folebridge mm. 114’ IS BWNl iNFF3(.! Fig. 2. — Geologic index map showing: (1) the location of the Kishenehn Basin and its setting with respect to other Tertiary basins of western Montana, and to the leading-edge of the Cordilleran fold- thrust belt (inset, upper right); and, (2) fossil localities of the North Fork region of the Kishenehn Basin. Numbered western Montana basins are (1) Kishenehn Basin, (2) Flint Creek Basin, (3) Deer Lodge Basin, (4) Smith River Basin, and (5) Three Forks Basin. Outcrops of Kishenehn Formation barren of invertebrate or vertebrate fossils shown as small gray dots and labeled with small-sized letters. Invertebrate localities, including some with vertebrate fossils shown as large gray dots and large-sized letters. Vertebrate localities shown as large open circles, large-sized letters and designated land mammal age. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 5 Table 1. — MoUuscan local faunas (If) of the mid-Tertiary Western Interior. Molluscan Local Fauna (l.f.) NALMA Map no. Reference Flint Creek l.f. Granite Co., MT e. Barstovian? 5 Pierce and Rasmussen, 1989 Cabbage Patch l.f. Granite and Powell Co’s., MT Arikareean 4 Pierce and Rasmussen, 1992 Pierce, 1992; 1993 Deep River l.f Meagher Co., MT Orellan- Arikareean 3 Roth and Embeton, 1994 Kishenehn l.f. Flathead Co., MT and S. Br. Col. White River l.f. Converse Co., WY Chadronian- Arikareean? Chadronian- Orellan 1 Russell, 1952; 1956 This paper Evanoff et ah, 1992 Three Forks l.f. Jefferson and Gallatin Co’s., MT Uintan- Chadronian 2 Roth, 1986 with only a small part of the basin. During our studies, a richly diverse mollusc an fauna, representing multiple climatic regimes, ranging from tropical to temperate, was collected from many localities exposed by both the North and Middle Forks of the Flathead River and their tributaries. The diversity was so great, both tax- onomically and environmentally, that a decision was made to report first on the generally younger(?) sediments and molluscan fauna found along the North Fork of the Flathead River. Even so, this paper deals with 55 molluscan taxa, of which 32 are new. This study focuses on the molluscan fauna recovered from ten localities in the North Kishenehn Basin along the North Fork of the Flathead River and its trib- utaries in Flathead County, Montana and eight localities in southern British Co- lumbia, Canada (Fig. 2). The southernmost Montana locality is along Bowman Creek (BWN-2), about 3.2 km above its confluence with the North Fork. The northernmost locality, in British Columbia, is along Commerce Creek (CMC-1), about one km above its confluence with the North Fork. With the exception of the Kintla Creek-Camegie (Montana) locality (KTL-1), the localities are all cut banks along the North Fork of the Flathead River or its tributaries. Five of the eight localities in British Columbia are at, or very near, Russell’s localities (1964: 541-542) and Russell’s locality designations are used informally, e.g., E-2 (NFF- 6), W-2 (NFF-5), E-3 (NFF-3), E-4 (NFE-2), and E-5 (NFF-1), without the im- plication that these are the identical localities studied by Russell. For location data and a description of each locality, refer to Appendix 1. The following abbreviations are used: CMNH or CM, Carnegie Museum of Natural History; KUMIP, University of Kansas Museum of Invertebrate Paleon- tology; p***, reference collection of H. G. Pierce; NALMA, North American Land Mammal “Age”; l.f., molluscan local fauna; Ma (Mega-annum), 10^ years before present. Geologic Setting. — The Kishenehn Basin is part of a network of late Paleogene grabens that formed during a post-Laramide episode of crustal extension that affected western North America from southern British Columbia to central Mex- ico (Coney, 1987; Dickinson, 1991; Constenius, 1996). This chain of grabens, of which the Kishenehn Basin is one of the northern-most, are superposed on the Cordilleran foreland fold and thrust belt, and formed when the thrust belt col- lapsed during a middle Eocene to early Miocene (ca. 49-20 Ma) episode of crustal 6 Annals of Carnegie Museum VOL. 70 Kishenehn Basin North Fork Flathead River - British Columbia North Fork Flathead River - Montana Fig. 3. — Geologic cross sections of the Kishenehn Basin showing relationship of basin stratigraphy to localities discussed in text. Note limited stratigraphic range of Kishenehn fossil localities. Sources of data include the following: Jones, 1969; McMechan, 1980; and, Constenius, 1981. Location of cross sections shown on Figure 2. extension. Consequently, extensional basins developed within this allochthonous terrain are rooted to preexisting thrust-fold structures and the sedimentary record of late Paleogene crustal spreading is found in thick, nonmarine, basin-fill assem- blages. These linkages are reflected in the modem Basin and Range landscape of southeastern British Columbia and western Montana in which the north to north- west-trending, elongate basins and congmous mountain ranges parallel the front of the fold-thrust belt (Figs. 1, 2). 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 7 The Kishenehn Basin, situated southwest of the Lewis Thrust salient, is a nar- row, asymmetric graben 150 km long and varies from 2 to 13 km wide (Con- stenius, 1982). Physiographically, the basin is surrounded by the rugged alpine peaks of the Lewis, Livingston and Clark ranges to the northeast, and by the comparatively lower relief MacDonald, Whitefish, Apgar-Belton Hills, and Flat- head ranges to the southwest. The basin is divided into two distinct regions, the North Fork and Middle Fork, by the Flathead River drainage. Cenozoic sedimen- tary rocks that fill this intermontane basin consist of two stratigraphic assemblag- es, late Paleogene rocks of the Kishenehn Formation and Quaternary glacial and alluvial sediments. The Flathead listric normal fault system, the master structure that controlled basin origin, borders the Kishenehn Basin to the northeast and is prominently defined in the landscape as the line of transition from subdued valley floor to precipitous mountain topography. Estimates of maximum dip-slip on the Flathead Fault are on the order of 15 km (Constenius, 1988). The southwest basin margin is variously bounded by normal faults antithetic to the Flathead system, such as the Nyack and Hornet faults, or it is onlapped by strata of the Kishenehn For- mation (Figs. 2, 4). Subsidence along these faults created an asymmetric graben containing up to 3,400 m (vertical thickness) of nonmarine, late Paleogene sedi- mentary rocks of the Kishenehn Formation. Because of the growth-fault sedi- mentary geometry of the basin, estimates of the maximum preserved stratigraphic thickness range from 4,750 m to more than 5,000 m (Jones, 1969; Constenius, 1981, 1988). Relict high-elevation basin-fill deposits indicate that at least 400 m (vertical thickness) of Kishenehn Formation has been removed by post middle Miocene erosion (Price, 1965; Jones, 1969; Constenius, 1981). The structural position of the Kishenehn Basin with respect to the Lewis Thrust salient reflects the common fault surface that accommodated late Cretaceous-early Eocene crustal contraction followed by middle Eocene-early Miocene extension (Figs. 1, 4, 5) (McMechan, 1981; Constenius, 1982). The linkage between the Lewis Thrust and Flathead Fault is seen in the position and symmetry of the fault systems. Details of the structural interactions of these linked contractile-exten- sional fault systems are reviewed by Bally et al. (1966), McMechan (1981), and Constenius (1982, 1988, and 1996). Importantly, the Flathead Fault has been identified as a southwest-dipping listric normal fault that merges with the reacti- vated segment of the Lewis Thrust. The fault dips —40-50° southwest near the surface and flattens at depth (Figs. 4, 5). Interpretation of seismic reflection data suggests that the Lewis-Flathead fault surface is layer-parallel in basal Cambrian strata. Consequently, the detachment horizon dips about 2-3° southwest parallel- ing an undeformed autochthonous crystalline basement (Bally et al., 1966; Yoos et al., 1991; Van der Velden and Cook, 1994). The synextensional nature of the Kishenehn Formation has been established using sedimentary structures, facies relationships, provenance, paleocurrent direc- tions, geophysical data, and stratal growth geometries (Constenius, 1981, 1982, 1988, 1989; McMechan and Price, 1980; McMechan, 1981). Synextensional sed- imentary sequences display stratal growth relationships, that is, a systematic thick- ening of strata toward the basin-bounding listric normal fault and a gradual flat- tening of dip in successively younger units (Dahlstrom, 1970; McMechan and Price, 1980). The strong rotational control on sedimentation associated with listric normal faulting results in a half graben or asymmetrical graben with a wedge- shaped sedimentary prism (Figs. 4, 5). The Kishenehn Basin is a half- or asym- 8 Annals of Carnegie Museum VOL. 70 §« Hiyi^ ?>2ir £»-i< 1=3 Q lydlz UJI^M TO< 1-iU.GQ is sg = VAON3M lAliU dnOMS NVMSZOa ^lyz = lygci) = ui9< = Q^S E 1 1 Uj Uj oc i 0 1 1 — m Z ifi la N N fa fa < Z < z < m H u O Q. UJ U o ly U O O U O Fig, 4, — Stratigraphic correlation chart of selected Tertiary basins in western Montana. Adapted from Pierce and Rasmussen, 1992a. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 9 metric graben in the central and northern parts of the basin where this study focused on invertebrate paleontology. The dip of the Kishenehn Formation strata in the North Fork region ranges from —35-40° in the lowest Duchesnean=age beds and decreases systematically upsection and toward the Flathead Fault to —20-25° (Fig. 2) (Constenius, 1981; McMechan, 1981). Hence, preserved in these middle Eocene to early Miocene synextensional strata is a record of 35-40° of total rotation of the hanging wall; —10-20° of growth-faulTrelated rotation, and —20-25° post-depositional rotation. Progressive rotation of beds indicates that displacement along the Flathead Fault system was synchronous with Kishenehn sedimentation. The distribution and age of Kishenehn strata preserved in the basin is not solely related to down-dropping and rotation on the basin-bounding faults. Rather, subtle warping of the Earth’s crust has imparted a regional northward tilt to the basin such that seismic mapping of the top of the crystalline basement shows consid- erable deepening to the north-northwest. Although the dip-rate is small, on the order of 1-2 degrees (17-34 m/km) over a distance of 100 km or more, the effect is large. As a consequence, middle Eocene (Uintan) rocks are exposed at the south end of the basin (Middle Fork region), in contrast to late Eocene (Duches- nean/Chadronian) rocks in the north end of the basin in Canada. Without this tilt we would expect a younger basin fill assemblage, perhaps Duchesnean/Chadron- ian age or younger, on top of the Pinchot Creek Member in the Middle Fork region (Fig. 3). The level of erosion in relation to regional arching of the basement is also expressed in the level of erosion of the fold-thrust belt. For example, in the hanging wall of the Lewis Thrust sheet, the youngest rocks preserved on the south end of the thrust salient are middle Proterozoic Belt Supergroup. Whereas on the north end in Canada, rocks as young as Devonian are preserved. Geo- chemical data on the thermal maturity of Kishenehn oil shales and coals provide direct evidence of a deeper level of erosion (i.e., more overburden removed) in the south end of the basin as compared to the north end (Curiale et al., 1988). Previous age determinations of the Kishenehn Formation concluded that the formation was late Eocene to early Oligocene in age (Russel, 1954, 1964; Hopkins and Sweet, 1976; McMechan, 1981). These studies relied on fossil mammals, mollusks, leaves, and pollen from exposures in Canada. Unpublished studies of palynomorphs by the U.S. Geological Survey suggested a late-middle(?) Eocene to late Eocene and late Eocene-01igocene(?) age for Kishenehn strata from the Middle Fork and North Fork regions, respectively (R. H. Tschudy, written com- munication, 1979; D. J. Nichols, written communication, 1980). In a reconnais- sance paleontologic study, Constenius et al. (1989) noted inconsistent paleonto- logic and geochronologic age estimates ranging from late Eocene to early Mio- cene. They postulated that the Kishenehn was deposited over a long time span and may have been a refugium with a mixture of faunal and flora types. Recent collection and study of mammalian fossils indicates that Kishenehn strata exposed in the North Fork region are mainly late Eocene-early Oligocene (Duchesnean-Arikareean; —41-24 Ma) in age (Figs. 2, 3), and a middle Eocene age (Uintan; —48-42 Ma) for Kishenehn strata in the Middle Fork region (M. R. Dawson, personal communication, 1998) is indicated. Discovery of a single large lobate leaf fossil of MacGinitea sp. in basal beds of the Kishenehn Formation suggests that part of the North Fork section may be as old as middle Eocene in age. Geochronologic analysis of two tephras yielded ^oAr/^^Ar ages ranging from 46.2±0.4 Ma for a sample from the Middle Eork region of the basin to 33.0± 1.0 Kishenehn Basin 10 Annals of Carnegie Museum VOL. 70 (SU9f 9Ui) llfdaQ (SJ9f9Ui) tf$d9Q 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 11 Ma for a sample from Commerce Creek in the North Fork region (Fig. 2) (Con- stenius, 1996; D. Archibald, written communication, 1997). Palynological analysis of a composite cuttings sample from the Cenex #4-13 Ladenburg well found a small number of late Paleocene-middle Eocene specimens in Kishenehn drill cut- tings (Bujak Davies Group, written communication, 1989). Strata of the Kishenehn Formation display a great variety of rock types, ranging from microlaminated oil shales and sapropelic coals to conglomerates and me- gabreccias (Price, 1965; Jones, 1969; McMechan and Price, 1980; Constenius, 1981; McMechan, 1981; Constenius and Dyni, 1983). Lithostratigraphically, the Kishenehn Formation has been divided into three parts, a Basal Member, a Lower Member, and an Upper Member. The contacts between these units are not exposed but are presumed to be conformable (Constenius, 1981; McMechan, 1981). How- ever, in other western Montana basins discrete basin-fill sequences are bounded by regional unconformities (Fig. 3) (Robinson, 1960; Rasmussen, 1973; Fields et ak, 1985; Hanneman and Wideman, 1991; Constenius, 1996). Additionally, Han- neman et ak, (1994) observed that most sequence-bounding unconformities are marked by paleosols or paleosol stacks. Well-developed paleosols are exposed at only three localities in the North Fork region. Trail Creek, Coal Creek, and Camas Creek, in lower Kishenehn strata (Constenius, 1981). These paleosols may be evidence of a sequence boundary between middle Eocene strata of the Basal Member and late Eocene-Oligocene Lower Member strata. Aside from the poten- tial sequence bounding, regional unconformity related, paleosols, the Kishenehn Formation has a demonstrable lack of mature paleosols and paleosol stacks that can be associated with long duration hiatuses in sedimentation. No angular un- conformities or intrabasinal normal faults have been identified on reflection seis- mic profiles across the basin (Fig. 5). Furthermore, the angular unconformity reported by MacKenzie (1916) has been shown by McMechan and Price (1980) to be the product of sedimentary lensing and channeling. The lack of condensed sections (paleosols), unconformities and faults with major offset, within the Kish- enehn sequence, combined with the affects of high sedimentation and subsidence rates, and the distribution of modem day exposures, allows us to make two im- portant assertions: 1) widely strike- separated fossil localities along the Flathead River drainage are approximately the same age, Duchesnean-Chadronian (Figs. 2, 4); and, 2) there is relatively little geologic time represented at each cutbank-type exposure along the Flathead River and its tributaries. The Basal Member in the North Fork region may be middle Eocene (Uintan) in age based on limited fossil data and, therefore, equivalent to Uintan strata of the Coal Creek and Pinchot members found in the Middle Fork region (Fig. 3). The Basal Member is a coarse-clastic, fluvial unit, consisting of yellow-tan and buff-colored conglomerate and conglomeratic sandstone with subordinate inter- beds of carbonaceous mudstone and coal, that rests above erosionally truncated Fig. 5. — Seismic reflection profile S-S’. (See Figure 2 for location). Image processing includes mi- gration to spatially restore reflections to their true position, time to depth conversion using velocity data from wells, and true-scale plotting (i.e., horizontal distance scale equals vertical depth scale). Notice that despite large-scale extensional movements on the Flathead fault (—10 km of displacement) that within the Kishenehn Formation layering is unbroken by normal faults (of significant size to be imaged on reflection seismic data) and there is no expression of intraformational angular unconfor- mities. Seismic data courtesy of Explor Data Ltd., Calgary, Alberta, Canada. 12 Annals of Carnegie Museum VOL. 70 Mesozoic, Paleozoic, and Precambrian rocks (McMechan, 1981; Constenius, 1981). Thickness estimates range from 140 m in British Columbia to 1,200 m in the west-central part of the basin in Montana (Fig. 4). In the Middle Fork region, middle Eocene rocks of the Kishenehn formation are a heterogeneous mixture of alluvial fan, fluvial, paludal, and lacustrine deposits that attain a composite thick- ness in excess of 1,500 m. Overlying the Basal Member are the dominantly fine-grained fluvial-lacustrine strata of the Lower Member. These strata are discontinuously exposed along the North Fork of the Flathead and its tributaries and it is from this unit that most of our fossil collections have been obtained. Based on analysis of associated fossil vertebrates, invertebrates, and microflora, this unit is late Eocene (Duchesnean- Chadronian) in age. Significantly, even though we’ve established 15 fossil local- ities within the Lower Member, they are clustered in the middle of the basin and confined to a limited stratigraphic age-range (Figs. 2, 4). Strata assigned to the Lower Member are an intercalated assemblage of light gray and light gray-green siltstones, mudstones, lignites, oil shales, marlstones, sandstones, and conglom- erates, and brick red and red-brown mudstones. The Lower Member attains a maximum thickness of about 2,000 m in Montana and more than 2,600 m in British Columbia (Jones, 1969:fig. 4). Detailed sedimentologic study of the Lower Member in British Columbia by McMechan (1981) identified four main depositional environments of a low relief alluvial-floodplain system: 1) braided river, represented by a range of coarse clastic deposits from horizontally laminated-massive sandstones to lenticular bod- ies of conglomeratic sandstone and conglomerate, 2) interfluve, characterized by massive to graded mudrocks, nodular marlstone, calcrete and gypsiferous mud- stone, 3) floodbasin pond and lake, characterized by fossiliferous marlstone and siltstone, carbonaceous mudstone, microlaminated shale, lignite, carbonaceous sandstone and rare sandy conglomerate, and 4) overbank near-channel floodplain, comprised of calcareous claystone and mudstone, fossiliferous marlstone, coalified tree roots and stumps, and sandstone and pebbly conglomerate. Southward, in Montana, widening of the basin corresponds with an increase in lacustrine and paludal deposits in the lower part of the Lower Member, and with an increase in reddish-oxidized fan-delta and floodbasin deposits in the upper part of the unit (Constenius, 1981). The Upper Member is a coarse-clastic unit composed of thick-bedded, boulder- cobble conglomerates, interbedded with sandstones, mudstones, and lignites. Pa- leocurrent, provenance, and other sedimentological data show that these are prox- imal alluvial fan and braided stream deposits derived from a rising mountainous terrain northeast of the Flathead Fault, the ancestral Clark and Livingston ranges (Price, 1965; Jones, 1969; McMechan, 1981). The presence of paleolandslide megabreccia deposits, that is, megaclasts of Cambrian and Devonian carbonates ranging from tens of meters to 1,800 m in length, provide evidence that there was significant paleotopographic relief across the Flathead Fault (Jones, 1969; McMechan, 1981). Dating of microfloral assemblages, a single vertebrate fossil (Merycoidodon sp.), and ^^Ar/^^Ar age of 33.0±L0 Ma establish an early Oligo- cene age for Upper Member deposits in British Columbia (McMechan, 1981; D. Archibald, written communication, 1997). Mammal fossils found in the Upper Member along Kintla Creek, Montana (Fig. 2) are early Oligocene-late Oligocene (Orellan-late Arikareean) suggesting that these are some of the youngest deposits 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 13 in the basin. The Upper Member is in excess of 2,100 m thick in British Columbia and is at least 1,500 m thick in Montana (Jones, 1969; Fig. 4). The Kishenehn Formation is overlain along a pronounced angular unconformity by a thin veneer (0-100 m) of late Neogene alluvial gravels and glacial detritus. No middle Miocene to Pliocene deposits have been identified in the Kishenehn Basin. There is little evidence of Quaternary-Recent movement of the basin- bounding faults with the exception of the mountain-front fault and associated triangular facets at Nyack Flats. Quaternary glaciation may have removed evi- dence of late-stage extensional faulting and sedimentation, but these products of tectonism were not of the same magnitude as in southwest Montana, where late Miocene to Recent basin-fill is lOO’s to 1,000’s of meters thick. Hence, subsequent to the middle Eocene-early Miocene episode of extension and sedimentation, the basin has been comparatively quiescent and was largely unaffected by Basin and Range extension ('^IT-O Ma). Methodology. — The North Kishenehn Basin is situated in rough, mountainous, heavily forested terrain. Roads are almost non-existent, being limited to a single, poorly maintained, gravel road originally linking Columbia Falls, Montana with Fernie, British Columbia along the west bank of the North Fork of the Flathead River. This road is now washed-out and abandoned in British Columbia. Since the road is currently impassable in British Columbia, access to the Canadian localities is now extremely difficult, requiring long, round-about approaches on narrow, heavily traveled logging roads. A second gravel road, in part nearly im- passable, is east of the River in Glacier National Park. The latter terminates at Kintla Creek, some distance short of the international boundary. These roads provide limited point access to tributaries of the North Fork of the Flathead River at crossings. Access to the river itself is limited to those few places where the “main” road and the river approach and the terrain is suitable. Elsewhere, land access is very difficult due to rugged terrain, heavy vegetation, and few game trails. Climatic conditions for the area are typified by Polebridge, Montana (Fig. 2) in the south end of the North Kishenehn Basin. The reported mean annual temperature (MAT) is 3.7°C with January and July means of — 8.5°C and 15.5°C. Annual precipitation is 57.2 cm, with more than half (57%) occurring principally as snow from October to March (www.worldwideweather.com). The typical fossil locality is a cut bank along the Flathead River (Fig. 6), with a few localities on major tributaries, that exposes sediments of the Lower Member of the Kishenehn Formation. Most of these outcrops have to be approached by rafting the river. Due to often long distances between raft launching and landing points, visits to a given outcrop locality were necessarily brief, generally two to six hours on a given outcrop, and were of a reconnaissance nature, which per- mitted only small grab samples of obviously fossiliferous zones, with obvious specimens picked directly from outcrop. On rare occasions, a single outcrop lo- cality would be subject to a near day-long visit in order to extract a significant mammalian specimen. Few localities were visited more than twice. In size, these outcrop localities varied from small, perhaps 3 m vertically above water level and perhaps 10-15 m in length, to quite large, over 30 m vertically and more than 100 m in length. Due to the near vertical nature of these cutbanks, collections were seldom made at heights greater than 3 or 4 m above water level. As a result, most localities were collected as a single unit without regard to local micro- stratigraphy. This appeared justified, since no evidence of significant faults or disconformity, other than the obvious unconformities between major units, e.g.. 14 Annals of Carnegie Museum VOL. 70 Fig. 6. — Typical cutbank style exposure of Kishenehn Formation along the North Fork of the Flathead River. The senior author and Alan Tabrum are shown collecting invertebrate fossils at the Wurtz Bend fossil locality (NFF19). Photo by K. Constenius, September, 1991. Basal and Lower members, and Upper Member to the overlying Quaternary till and alluvium, were noted at the various localities. The fossiliferous sediments vary from silty shales and marls to fine, usually silty sandstones. Most molluscan fossils are badly crushed due to overburden compression. Disaggregation of the grab samples in water results in a hash of nearly unidentifiable fragments. Successful extraction of specimens requires sta- bilization upon exposure with a dilute solution of poly-vinyl acetate (PVA), then subsequent removal of additional matrix, softened by water, using dental tools and/or needle probes and additional PVA treatment. Most specimens required repeated steps to extract completely or to expose sufficiently for study. Some specimens were exposed only far enough to reveal significant characteristics, then, to avoid inadvertent breakage, left on small blocks of matrix. Systematic Paleontology Terrestrial Taxa Order Archaeogastropoda Thiele, 1925 Superfamily Helicinoidea Latreille, 1825 Family Ceresidae Thompson, 1980 Type genus. — Ceres Gray, 1856. Environment and Distribution. — The Ceresidae are known only from Mexico 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 15 and South America and contain five modern genera (Thompson, 1980:16-17). They are found in warm, moist, forested, tropical environments, often on a lime- stone substrate. The genus Ceres is endemic to eastern Mexico, in the states of Tamaulipis, San Luis Potosi, and Vera Cruz. Geologic range of the family is unclear. Other than genus Tozerpina, which ranges from Cretaceous to Oligocene in Alberta and Montana, I know of no other reference to this family as fossil. Genus Tozerpina Bishop, 1980 Type species. — Tozerpina rutherfordi (Russell, 1929). Discussion. — Although Bishop (1980:243, 245) clearly described the genus, and listed three included species, he did not designate a type species from among the three. Tozerpina rutherfordi is herein so designated, based on priority of publication of species, and as most typical of Bishop’s Diagnosis of the Genus. The other two species included by Bishop are T. douglasi (Tozer, 1956) and T. mokowanensis (Tozer, 1956). The taxa following are hereby included and become the fourth, fifth, and sixth taxa of this genus, all fossil, extending the geologic range of the genus from latest Cretaceous into late(?) Eocene, and the size range down to 2.5 mm in width. Among the Ceresidae, which have great variety in apertural lamella, both in number and location, Tozerpina, with three lamellae, two palatal and one basal, appears most similar to the type genus, Ceres, which may also be keeled and the only modem genus in the family with palatal lamellae. Tozerpina buttsi (Russell, 1956), new combination (Fig. 7.A-7.C) Triodopsis buttsi Russell, 1956<3:108, 109, figure 3, plate I, figures 1-6; 1964:538, 541; Ross, 1959: 70, 71; Roth, 1986:table 2; Constenius et ah, 1989:table 2. Discussion. — Russell’s (1956:108, 109, fig. 3, pl.l, figs. 1-6) type specimens were examined to determine that this species rightfully belongs to the genus Tozerpina as defined by Bishop (1980:243, 245). They differed from the type description only in possessing a peripheral keel. Russell (1956:109) reported "'Triodopsis'' buttsi from only one locality, his E-4. His exact locality has been subsequently destroyed by erosion and construction of a highway bridge. Al- though there are molluscan bearing sediments at the approximate locality, no identifiable Tozerpina were among the taxa recovered there, although fragments recovered may represent one or more species of the genus. Figures 7.A-7.C are of Russell’s holotype (1956:106), Geological Survey of Canada (GSC) 11606, diameter 4.5 mm, height (crushed, estimated) 2.25 mm, 5 whorls. Tozerpina buttsi parva Pierce, new subspecies (Fig. 7.D-7.I) Diagnosis. — A small Tozerpina buttsi, nearly identical in all proportions, but with slightly stronger and more closely spaced collabral costellae. Description. — Shell tiny, lenticular, of about 4.50 to 4.75 whorls, inner walls of initial whorls resorbed, periphery modestly carinate, suture scarcely impressed. Embryonic whorls about 1.5, finely granulose, later whorls with fine, very closely spaced collabral costellae, about 20 per circumferential millimeter, initially faint, becoming stronger on later whorls, costellae continue over periphery, be- coming somewhat weaker, but continue to margin of a large callus pad (Vs diameter) that conceals the umbilicus. Aperture ovate-lunate, retractive at about 45°, with three teeth, outer lip strongly thickened and reflected, slightly sinuous above periphery, edge rebated by rounded groove extending from suture to umbilicus. Upper palatal tooth located midway between periphery and upper insertion, initially wide Table 2,—Kishenehn molluscan fauna, terrestrial, by locality. X indicates present at locality. X indicates type locality, or locality of figured specimen. Entry in columns to the right of BWN-2 if taxon is previously known; CPch = Cabbage Patch; 3Fks = Three Forks; RSL = Russell’s Kishenenh fauna. 16 Annals of Carnegie Museum VOL. 70 p Is feu feg 2c^ Tz feu zS few z"" feS yd X X X X X X XX XX XX XXX 3 . s m W ^ . pq K X U 6^ s « § •2 i ^ -2 u G s s X X m S Q „ ^ o < -S o p ^ X X XX X X X X X X X X X X X XXX X X X XX X XXX X X X X X X X X X X XX X XXX X X X X X X X X X X X p c- K X w s " L w S s S s U I d D P P P 00 p ® I 5 -p § ^ s s|^ a 3 s I i t>5 i.. I ^ s 6 5 g © w I •S s ”5 -2 5 < ^ Holospira tabrumi n.s. Holospira beardi n.s. Coelostemma dawsonae e.s. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 17 I u m z ! M ,U feu p few yd X XX X X X X X X X XX X X X X X X X X X X X X s uJ .Si O ii, ^I>1 S 5 §J :5 s Q X ^ <« o o s s Q Q s "S a, e I g I o < N 5 s d, »>« ca Q Q U s § a -a 11 a I s 5^ d X II O X X X X Q a •« ^ I a: 5 X s'? o o X X a .s ^ X ^ 5 < I o H i O Q ffi X ^ ' o ^ m g 3 ^ w XerarionaJa constenii n.s„ 18 Annals of Carnegie Museum VOL. 70 Fig. 7.— ARCHAEGASTROPODA; CERESIDAE. A-C: Tozerpina buttsi (Russell, 1956), Holotype (GSC 11606), XIO; A. apical; B. umbilical; C. apertural; D-F: Tozerpina buttsi parx’a n. ssp., X16; D-E. Holotype (CM 41614); D. dorsal; E. umbilical; E Paratype (KUMIP 289700), aperture; G-I; reconstruction XI 6; G. lateral; H. apertural; I. dorsal. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 19 and blunt, becoming narrower and sharply crested, deeply entering, slanting downward, causing de- scending, deeply impressed dorsal sinus on exterior of shell behind lip, the outer margin of which rises to become the peripheral carina. Lower palatal tooth located about midway between periphery and columella, narrower and higher, spirally entering, crest deflected adaxially with rounded groove below crest, may develop secondary node on abaxial side shortly after entering, forms short, shallow sulcus below and behind the lip, palatal teeth form prominent oval sulcus on inside of peripheral margin of lip, which merges with the descending, sharply rounded ridge of the dorsal plica shortly behind lip. Basal tooth very close to columella, fine, shorter, crest deflected abaxially, spirally entering, separated from lower palatal tooth by small, rounded sulcus. Parietal callus strong, margin curved, joining with heavy basal callus. Measurements of holotype: width, major 2.5 mm; minor 2.3 mm; height 1.45 mm; HAV 0.58; about 4.5 whorls. Ranges: width, major 2.5 to 3.0 mm; minor 2.3 to 2.75 mm; height 1.45 to 1.55 mm, most crushed; 4.5 to 4.75 whorls. Discussion. — Tozerpina buttsi parva is, quite simply, a miniature T. buttsi, proportionally smaller in all measurements, and is not simply a less mature spec- imen. Tozerpina buttsi buttsi, at 4.5 to 4.75 whorls, typical of mature T. buttsi parva, is still much larger in all dimensions. Tozerpina buttsi parva also differs from the nominate form in having slightly stronger, and more closely spaced, collabral costellae. Tozerpina lends, described subsequently herein, differs from T. buttsi parva in being much larger, lacking a peripheral carina, and having coarser costellate. Tozerpina buttsi parva is, by far, the smallest species of its genus and one of the smallest of its family. Tozerpina rutherfordi (Russell 1929) differs from T. buttsi parva in being much larger, width 9 mm, in having two parietal lamellae, a deep set palatal lamellae beginning well behind the lip, and six costellae per millimeter on the ultimate whorl. Tozerpina douglasi (Tozer, 1956) also differs in size, 10 mm, and in having spiral as well as radial sculpture with 15 radial costellae per millimeter on the ultimate whorl. The Cretaceous species, T. mokowanensis (Tozer, 1956), width of 5 mm, has 20 costellae per millimeter on the ultimate whorl, and a keeled pe- riphery, and most closely approximates T. buttsi parva. It, too, is larger, and differs in having only one palatal lamella, and a body whorl that descends slightly in the last millimeter. Tozerpina buttsi and T. mokowanensis are the only species of this genus with a keeled periphery. Etymology. — Latin parva, small or little. Material. — More than 30 specimens. Holotype CM 41614, and paratypes CM 41615 (2), KUMIP 289,700, KUMIP 289,701 (2), PCER-104 (10 fragments), and PCER-105 (2 molds w/plastitypes) are from Commerce Creek (CMC-1) locality, British Columbia. Paratypes: PCER-101 (2), PCER-102 (10 fragments), PCER-103, and PCER-106 (2) are from E-5 (NFF-1) locality, British Columbia. All spec- imens from the Lower Member, Kisheeehn Formation, late Eocene. Tozerpina lends Pierce, new species (Fig. 8.A-8.E) Diagnosis. — A small, lenticular Tozerpina of about five whorls, with about six collabral costae per circumferential millimeter at periphery of the last whorl, and a heavily strengthened retractive aperture containing three teeth. Description. — Shell small, lenticular, 4.75 to 5.25 whorls, inner walls of initial whorls resorbed, periphery roundly acute but not keeled, suture scarcely impressed. Embryonic whorls about 1.75, finely granulose, later whorls with about six regular collabral costae per circumferential millimeter, initially weak, becoming very distinct on later whorls, weakening slightly after passing over periphery, but continuing until covered by large callus pad that conceals umbilicus. Aperture ovate-lunate, re- tractive at about 45°, outer lip strongly thickened and reflected. Teeth three, upper palatal tooth located nearer insertion, initially blunt, deeply entering, becoming narrower and more sharply crested, curving slightly abaxially, causing impressed sulcus on upper lip, lower palatal tooth located midway between periphery and insertion, narrower and higher, entering spirally, crest deflected adaxially, basal tooth 20 Annals of Carnegie Museum VOL. 70 Fig. 8.— ARCHAEGASTROPODA: CERESIDAE. A-E: Tozerpina lentis n. sp., XIO; A-B; Holotype (CM41616); A. apical; B. umbilical; C. Paratype (CM 41617), XIO, lateral, laterally compressed; D- 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 21 small, very close to columella, crest deflected abaxially, spirally entering, separated from lower palatal tooth by rounded sulcus. Parietal callus thick, margin curved to join thick basal callus. Measurements of holotype: width, major 4.6 mm; width, minor 4.0 mm; height about 2.5 mm, crushed; 5 whorls. Ranges: width, major 4.0 to 4.8 mm; width, minor, 3.5 to 4.4 mm; height, ?, all specimens crushed; whorls 4.5 to 5.25. Discussion. — Tozerpina ientis most resembles T. buttsi buttsi, but, although of about the same size, it differs in having a coarser sculpture (—six costae versus 15-20 costellae per mm), an acutely rounded periphery, and in the position of the upper palatal tooth and the associated external sulcus behind the outer lip. Among the older Cretaceous-Paleocene taxa of the genus, it shares the charac- teristic of three palatal lamellae with T. douglasi and of six costae per mm with T. rutherfordi, but differs from the former in size, and in lacking spiral sculpture, and from the latter in size and in lacking parietal lamellae. Fragments found at Locality E-5 (NFF-1) may represent T. Ientis. Etymology. — Latin Ientis, lentil, lens=shaped. Material. — Nine specimens. Holotype CM-41616, and paratypes CM“41617 (2), KUMIP 289,702 (2), PCER-107 (2), and PCER-108, are from Kintla Creek/River (NFF-17) locality, Flathead County, Montana, eight specimens. Type horizon: Lower Member, Kishenehn Formation, late Eocene. PCER- 109 is from E-5 (NFF=1) locality, British Columbia, Lower Member, Kishenehn Formation, late Eo- cene. Family Helicinidae Latreille, 1825 Type genus. — Helicina Lamarck, 1799. Subfamily Hendersoninae Baker, 1926 Type genus.- — Hendersonia A. J. Wagner, 1905. Environment and Distribution.— Hendersonia are “found on well-shad- ed, leafy, and rather humid slopes, on limestone terranes” and are spottily dis- tributed “in the upper Mississippi Valley east to the Alleghenies in Pennsylvania and North Carolina” (LaRocque, 1970:555-556). Recent Waldemaria live only in Japan, although it is known as fossil from middle Miocene (early Barstovian) Flint Creek beds of southwestern Montana (Pierce and Rasmussen, 1989), where there is a suggestion of a moist, temperate climate. Genus Waldemaria Wagner, 1905 Type species. — Tomichia japonica (Adams, 1861:308). Waldemaria monticula Pierce, 1989 (Fig. 8.F-8.I) Discussion.^ — Except for minor differences such as the body whorl being slight- ly more rounded than the holotype, which could easily fall within the range of normal specific variation, and the presence of low obscure costae (about six per circumferential millimeter) overprinted on the fine collabral costellae (>20 per f- E: restoration; D. lateral; E. apertural; HELICINIDAE. F-I: Waldemaria monticula Pierce, 1989 (in Pierce and Rasmussen, 1989), F-H: Hypotype (CM 41622), X5; F, dorsal; G. umbilical; H. apertural; I: W. monticula, Hypotype (PHCN 117), XIO, partial operculum; J-M; Lucidella columbiana n. sp., J-L: Holotype (CM 41620), X5; J. apical; K. umbilical; L. apertural; M. L. columbiana Paratype (KLfMIP 289705) XIO, apical detail, apex to lip; N-R: Lucidella salishora n. sp., N-Q: Holotype (CM 41618), X5; N. apical; O. umbilical; P. apertural; Q. lateral; R. L. salishora, Paratype (KUMIP 289704), XIO, apical detail, apex to lip. 22 Annals of Carnegie Museum VOL. 70 millimeter) on the last two whorls, which may have been worn off the holotype, there are no significant differences between these specimens and the holotype. A partial Waldemaria, cf., W. monticula operculum was recovered at locality CMC- 1 (Figure 8.1) although not in direct association with the single crushed specimen from that location. This operculum, although incomplete, compares almost per- fectly with the operculum of the holotype (Pierce and Rasmussen, 1989: fig. 2-2, 2-5). There is no hesitation in identifying these specimens as W. monticola. Mea- surements of hypotype: width, major 9.2 mm; width, minor 8.5 mm; height 8.0 mm; HfW 0.87; 4.75 whorls. Material. — Ten specimens. Figured specimen CM-41622 and hypotypes CM-41623, PHCN-114 (2) are from Lower Kintla Rapids (NFF-18) locality. Additional hypotypes: KUMIP 289,707 is from Wurtz Bend (NFF-19) locality; PHCN-115 (3) is from E-4 (NFF-2) locality; PHCN-116 is from Commerce Creek (CMC-1) locality (1), both British Columbia, all from Lower Member, Kishenehn Formation, late Eocene. A partial operculum of Waldemaria, cf. W. monticula, PHCN-117 was recovered at the Commerce Creek (CMC-1) locality, British Columbia, same horizon. Subfamily Helicininae Latreille, 1825 Type genus. — Helicina Lamarck, 1799. Genus Lucidella Swainson, 1840 Type species. — Helix aureola Ferussac, 1822. Subgenus Poeniella H. B. Baker, 1923 Type species. — Helicina (Plicatula) christophori Pilsbry, 1897 (fide Pilsbry, 1948:1085). Discussion. — The subgenus Poeniella is distinguished by axial riblets with or without spiral striae. Environment and Distribution. — The genus Lucidella is found in warm, moist tropical environments, usually well wooded, and is found throughout the Greater and Lesser Antilles, and circum-Caribbean from Venezuela to Yucatan (Thiele, 1992:127), and southern Vera Cruz (Baker, 1922:36). Although the subgenus Poeniella is most diverse on Hispaniola (Pilsbry, 1928:479), there are several species on Cuba (Aguayo y Jaume, 1947-1951:395), one of which has also been found in the Florida Keys (Pilsbry, 1949:1085). Geologic range of the subgenus is unknown. Pre-Quaternary fossils of the genus Lucidella are almost unknown. I have examined the type material for “Helicina'’ oregona Hanna 1920, from the Oligocene John Day beds of Oregon, and find that it is, unquestionably, a Luci- della, intermediate in size between the two new species described herein. Good- friend and Mitterer (1988:301) reported two extant species as latest Pleistocene fossils from Jamaica. Lucidella Columbiana Pierce, new species (Fig. 8.J-8.M) Diagnosis. — A large Poeniella with sinuous collabral costae crossing micro- scopically faint spiral costellae, noticeably constricted behind dorsal margin of outer lip, aperture reflected and reinforced but without teeth. Description.Sh&\\ large, depressed conic globose, whorls 4.75, umbilicus covered by callus pad, inner partitions of spire resorbed, sutures scarcely impressed. Embryonic whorls about one, very finely and irregularly punctate, nucleus 0.45 millimeter in diameter, subsequent two whorls finely and faintly costellate, costellae collabral, slightly sinuous, superimposed on very faint spiral costellae that weaken on ultimate whorls, but still visible under high magnification, especially on base, last 1.75 whorls costate, about seven per circumferential millimeter, continue over rounded periphery only slightly 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 23 diminished until covered by umbilical callus. Aperture lunate-ovate descending in last millimeter, retractive from axis about 45°, outer lip reflected and reinforced moderately, narrowly constricted behind dorsal margin of outer lip, parietal callus extends to smooth to finely granulose callus pad hiding umbilicus, umbilical callus about Vs diameter of shell. Measurements of holotype: width, major, 9.4 mm; width, minor, 9.0 mm; height 5.2 mm; whorls 4.75; HAV 0.55. Discussion. — Lucidella columbiana resembles both “Helicina” oregona and Lucidella salishora. ‘'Helicina” oregona is smaller, has more whorls, 6 versus 4.75, taller, HAV 0.625 versus 0.55, is more finely costate, 9-10 versus 7 per millimeter, with the costae fading nearly completely on the base, and it has a basal tooth in the more prominently descending aperture. Lucidella columbiana appears to be most similar to L. salishora, which differs in having more whorls with angular juvenile whorls, being much smaller, having a strong thickening of the upper portion of the outer lip, and a narrow, deep constriction behind the entire outer lip. Etymology. — Named with reference to the Columbia River, of which the Flathead River is a trib- utary. Material. — Fifteen specimens. Holotype, CM-41620, and paratypes: CM-41621 (2); KUMIP 289,705; KUMIP 289,706 (2); PHCN-110; and PHCN-111 are from E-5 (NFF-1) locality, British Columbia, Lower Member of the Kishenehn Formation, late Eocene. Additional specimens: Commerce Creek (CMC-1) locality, PHCN-1 143(6); locality data as above, and Wurtz Bend (NFF-19) locality, PHCN-112, Flathead County, Montana, both localities in Lower Member, Kishenehn Formation, late Eocene. Lucidella salishora Pierce, new species (Fig. 8.N-8.R) Diagnosis. — A moderately large Poeniella with sinuous collabral costae cross- ing microscopically faint spiral costellae, noticeably constricted behind outer lip, which is strongly thickened on dorsal margin, without basal tooth. Description. — Shell of moderate size, conic globose, whorls 5.25-5.50, inner partitions of spire resorbed, sutures scarcely impressed. Embryonic whorls about 1.25, finely punctate, nucleus 0.3 mm in diameter, neanic whorls about 0.75 with very fine collabral costellae, which become prominent and more widely spaced sinuous collabral costae, about six per circumferential millimeter, on last three whorls, continue over roundly angular periphery, diminishing somewhat on base, crossing faint spiral costellae until covered by callus. Initial four whorls distinctly angular at periphery. Aperture roundly lunate externally, descending slightly in last millimeter, very noticeable narrow constriction immedi- ately behind outer lip, outer lip reflected and roundly thickened to an acute margin, retractive from axis almost 60°, interior margin of upper part of outer lip has wide, low, blunt reinforcement over % of distance from point of insertion to periphery, about twice as thick as remainder of lip when viewed aperturally, giving inner margin of aperture a roundly angular shape. Base slightly sunken, covered by smooth to faintly granulose callus pad that merges with parietal callus. Measurements of holotype: width, major, 5.0 mm; width, minor, 4.2 mm; height 2.5 mm; HAV 0.5; whorls 5.25. Measurement of aperture: width 1.2 mm, height 1.7 mm. Range of measurements: width, major, 3. 6-5. 7 mm; width minor, 3. 0-4. 6 mm; height, 2. 3-3. 3 mm; HAV 0.50-0.64; whorls, 5.25-5.50. Discussion. — The “axial, instead of spiral riblets” (Baker, 1923, 9.23) clearly place Lucidella salishora in the subgenus Poeniella. It rather closely resembles modern Cuban Lucidella rugosa (Pfeiffer, 1839) and Cuban and Floridian LucL della tantilla (Pilsbry, 1902), having similar sculpture, HAV ratios, and in lacking a basal tooth. It differs from the aforementioned primarily in size and in the coarseness of the sinuous collabral costellae. The Oligocene ‘'Helicina’* oregona is similar in size and is noticeably constricted behind the aperture, especially dorsally. It differs from L. salishora in being slightly larger, taller, HAV 0.62 versus 0.50, more finely costate, 9-10 versus six per millimeter, has an aperture 24 Annals of Carnegie Museum VOL. 70 that descends more noticeably, lacks the thickening of the dorsal portion of the outer lip, and has a blunt basal tooth separated from the columella by a rather deep notch. Etymology. — Named to honor the Salish speaking Flathead Indians. Material. — More than 150 specimens. Holotype, CM-41618, and paratypes: CM-41619 (3); KUMIP 289,703; KUMIP 289,704 (3); PHCN-103-106 (63), PHCN-109 (10) are all from Commerce Creek (CMC-1) locality, British Columbia (>100 specimens). Type horizon is the Lower Member, Kishenehn Formation, late Eocene. Additional localities are as follows: E-5 (NFF-1), PHCN-107 (23), and E-4 (NFF-2), PHCN-108 (15), both British Columbia, horizon as above. Order Stylommatophora Schmidt, 1835 Superfamily Succineoidea Beck, 1837 Family Succineidae Beck, 1837 Type genus. — Succinea Draparnaud, 1801. Genus Oxyloma Westerlund, 1885 Type species. — Succinea dunkeri Pfeiffer, 1849. Oxylomal kintlana Pierce, new species (Fig. 9.A) Diagnosis. — Shell narrow with long acute spire, and narrowly ovate-lunate ap- erture about as tall as the spire. Description. — Shell narrow, of more than three rounded, expanding whorls, suture well impressed, spire long, acute, nucleus of embryonic whorls elevated, acute, very finely costulate, remaining whorls with sculpture of fine, closely packed, orthocline growth wrinkles. Aperture narrowly ovate-lunate, about equal to spire height, peristome simple, sharp. Measurements of figured holotype, CM 41628: height (est.), >8 mm; width (est.), 4 mm; height of aperture, 4 mm; whorls (est.), >3. Discussion. — All specimens are badly crushed and incomplete, but all features except the columella can be seen on one or more of the specimens. Under the best of circumstances, it is next to impossible to identify succineids from shell alone, except for the genus Oxyloma. The condition of the specimens at hand precludes precise identification of even that genus. Overall character of this spe- cies does, however, strongly suggest that these specimens are, indeed, Oxyloma and, most probably, in the O. ejfusa Group (Section Neoxyloma Pilsbry) (Pilsbry, 1948:777). This species is very different from late Oligocene (Succineidae) mon- tana Pierce, 1992, and all other succineids reported from pre-Pliocene strata (Gal- breath, 1969:91-94, fig. Id; White, 1883:45,46, pi. 19, fig. 4a). Etymology. — kintlana, with reference to the type location along Kintla Creek, a west flowing stream entering the Flathead River from Glacier National Park, Montana. Environment and Distribution. — The O. ejfusa Group is North American, gen- erally north of the 35th parallel, but extends into Florida, and is essentially absent from the Southwest and Far West of the U. S. They are normally encountered in well-vegetated, permanently marshy to swampy localities (LaRocque, 1970:697- 699). Geologic range of the succineids is not well understood, partially due to the fragile nature of the shell, and to the intergrading shell morphology among species, but is probably of great antiquity, based on their anatomy (Patterson, 1971). The genus Oxyloma is fairly common in late Pliocene (Blancan) faunas on the High Plains. The late Oligocene (Succineidae) montana Pierce, 1992 may be a Catinella. Galbreath (1969) found what could well be succineids in the Oligo- Miocene of northeastern Colorado, but, being molds, precise identification is not 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 25 possible. Eocene ‘*Succinea” papillispira White, 1876, may not be a succineid (Pierce, 1992:611). Material. — Four specimens. Holotype and figured specimen, CM-41628 and paratypes CM 41629 (3), are from Kintla Creek River (NFF-17) locality, Flathead County, Montana. Two unidentifiable succineid spire fragments from South Ford Creek (NFF-22) locality, PSUC-105, may represent this species. All localities in Flathead County, Montana, Lower Member, Kishenehn Formation, late Eo- cene-Oligocene. Genus Omalonyx Orbigny, 1841 Type species.— Succinea unguis Martens, 1873:193. Discussion. — -This is a slug-like genus of the Succineinae (^. str.) that currently inhabits Central and South America and the Antilles. Thiele (1992:800-801, fig- ure 587), using an alternative spelling of Homalonyx, describes the genus as: “Shell flatly bulging, fingernail shaped, with very small terminal spire and very wide, elongated oval aperture, covered by the mantle along the margin or com- pletely.” Patterson (1971:147-151; 181; 187; 193-195, figures 38-46) has studied O. felina Guppy, 1872, and amplifies the shell description as follows: “The shell of the largest individual measures 11.5 mm in length 7.3 mm in width with an aperture length of 1 1 mm (Fig. 38). The shell is very depressed in side view and has one Va whorls. There is no spire. The aperture occupies almost the entire shell, and is oval in shape. There is no columellar plait but there is a noticeable colu- mellar flange which projects into the aperture. The shell is very thin, transparent and light amber in color. The coloration is slightly darker at the apex. The shell is sculptured with fine, readily visible growth lines and some widely spaced radial lines which can be seen only with transmitted light. The shell has a glossy ap- pearance inside and out. The edge of the mantle encircles the peripheral 2 mm of the shell.” Environment and Distribution. — '"Omalonyx is a slug-like succineid inhabiting tropical Central and South America and various islands of the West Indies. There are only a few recognized species and all are amphibious in habit, living near or on emerging vegetation of bodies of freshwater” (Patterson 1971:181). Geologic range is unknown. Omalonyx antiqua (Russell, 1956a), new combination (Fig. 9.B, 9.C) Binneya antiqua Russell: 1956<3, p. 112-113, fig. 7, pi. 3, figs. 7-10 Description. — “Shell small, oval, low, consisting of about one whorl which expands rapidly to form a broad, shallow body cavity, with the posterior margin somewhat turned in ventrally as part of a vestigial spire. Outline more convex on left side than on right. Surface almost smooth, with occasional faint growth lines. Length of holotype, 5.7 mm; width 3.8 mm” (Russell, 1956<3: 1 12-1 13). Discussion. — It has been recognized for some time that there is a serious prob- lem with the taxonomic position of "‘Binneya’" antiqua Russell, 1956a. Roth (personal conununication, 1990) early recognized that assigning this species to the genus Binneya Cooper, 1863 was inappropriate. Comparison with specimens of Binneya notabalis Cooper, 1863, kindly provided by Roth, quickly confirmed this. The shape of the shell provided strong evidence that the animal had been slug-like, but attempts to fit “B.” antiqua into the other slug-like families were less than satisfactory. The Limacidae provided a posssible fit, but it, too, was far from satisfying. The hard, calcific shell of “B.” antiqua, originating from a spire Annals of Carnegie Museum VOL. 70 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 27 of 0.25 to 0.50 whorl, differs markedly from the normal character of limacid slug plates (shells). The shell of “R '' antiqua does compare well with the descriptions of the genus Omalonyx in almost all ways. The only difference noted is the shell is slightly thicker that described for O. felina. The net result is to remove ''B.” antiqua from the genus Binneya and place it within the succineid genus Omalo- nyx, a much more satisfactory taxonomic position. Omalonyx cocleare Pierce, new species (Fig. 9.D-»9.F) Diagnosis.— A small, spoon-shaped shell, embryonic shell with 0.25 to 0.50 dextral whorls before elongate aperture nearly as tall as shell. Description. — Shell small, substantial, elongate, gently rounded, inverted spoon shaped, convex up, with smooth, shiny, solid, white dorsal inductura, basal concavity (aperture) nearly as large as shell, containing yellow, crystalline calcite inner layer, partially to nearly filling the concavity; embryonic shell begins with about 0.25 to 0.50 dextral whorls before opening into an elongated, ventrally located, aperture, subsequent growth elongate and widening to attain an oval, spoon shape with a bluntly rounded distal margin, protoconch smooth, slightly protruding from posterior margin of shell with apex well curved to right, distinct hook shape ventrally; convex dorsal surface of shell sculptured with weak, closely spaced, semicircular growth costellae, with one or more costellae somewhat stron- ger. Measurements of holotype, CM 41630: height (length) 3.8 mm; width 2.2 mm; thickness 0.7 mm; whorls about 0.35. “This taxon was first thought to be a form of Omalonyx antiqua^ which Russell (1956:112-113, pi. 3, figs. 7-10) had collected from his nearby locality E-4. Subsequent evaluation under the microscope showed this taxon to be distinctly different, with less than 0.50 whorl coiling in the embryonic stage before growth becomes linear. In O. antiqua^ the initial spiral stage lasts one, or slightly more, whorls. It should be noted that, despite careful collection of our E- 4 locality, no specimens of O. antiqua were recovered. However, the locality collected as E-4 was not, in fact, exactly the same as RusselFs early 1950’s locality. Flood erosion, and construction of a modem highway bridge, have se- verely modified the local terrain. All other taxa listed by Russell (1956:105) from his Locality E-4, except Tozerpina buttsi (Triodopsis buttsi) (Russell, 1952), were collected. Comparison with Deroceras pachyostracon (Taylor, 1954) from the Pliocene of California, which also has a short initial spiral stage of about 0. 1 whorl, showed major differences in the character of the inductura. The inductura is hard, dense, relatively thick and shiny for O. cocleare, versus chalky, thin and dull for D. pachyostracon, which also has a much thicker inner yellow calcitic layer. Fig. 9.— STYLOMMATOPHORA: SUCCINEIDAE. A: Oxylomal kintlaense n. sp. holotype (CM 41628), X5, apertural; B-C: Omalonyx antigua (Russell, 1956) Holotype (GSC 11621), X5; B. dorsal; C. ventral; D-F: Omalonyx cochleare n. sp. Holotype (CM 41630), XIO; D. dorsal; E. ventral; F. lateral; HAPLOTREMATIDAE. G-H: Haplotrema simplex (Russell, 1956) X3; G. Hypotype (CM 41653), apical; FI. Hypotype (PHPL-102) umbilical. LIMACID AE. I-J: Deroceras securis Pierce, 1992 Hypotype (CM41663), X5; I. dorsal; J. lateral; K-L: Deroceras mahiz Pierce, 1992 Hypotype (CM 41665), X5; K. dorsal; L. lateral. DISCIDAE. M. Discus mackenziei Russell, 1956 Hypotype (KUMIP 289733), XIO; N-P: Anguispira sp., c.f. A. alternata (Say, 1816); N. Hypotype (CM 41661), X3, apertural; O-P: Hypotype (KUMIP 289734), X5; O. apical; P. umbilical. ZONITIDAE. Q-R: Nesov- itrea pulchra Pierce, 1992 Hypotype (CM 41667), XIO; Q. apical; R. umbilical. 28 Annals of Carnegie Museum VOL. 70 Etymology. — cocleare, Latin, a spoon, a reference to the shape of the shell. Material. — Twenty specimens. Holotype, CM-41630, and paratypes: CM-41631 (3); KUMIP 289,710; KUMIP 289,711 (3); PSUC-106 (2), and PSUC-107 (5) are from Commerce Creek (CMC- 1) locality; additional paratype PSUC-108 (5) is from E-5 (NFF-1) locality, both British Columbia, Lower Member, Kishenehn Formation, late Focene-Oligocene. Superfamily Pupilloidea Turton, 1831 Family Pupillidae Turton, 1831 Subfamily Pupillinae Turton, 1831 Genus Pupoides Pfeiffer, 1854 Type species. — Bulimus nitidulus Pfeiffer, 1839, by subsequent designation (Kobelt, 1880). Subgenus Ischnopupoides Pilsbry, 1926 Type species. — Pupa hordacea Gabb, 1866. Environment and Distribution. — Characteristic of dry country, essentially the Rocky Mountain states of western United States, from southwest South Dakota through Arizona to the state of Sinaloa, Mexico, with an outlying species in western South America (Pilsbry, 1948:924; Bequaert and Miller, 1973:58-59). Pupoides hordaceus, which is similar to the following species, is a dry country species, generally inhabiting the Upper Sonoran Life Zone of the southern Rocky Mountains, from southeastern Wyoming to New Mexico and all of Arizona but not, apparently, found in adjacent Sonora, Mexico (Bequaert and Miller, 1973: 58-59, 177). Known geologic range of Ischnopupoides is from late Eocene-early Oligocene (Chadronian) {P. tephrodes Roth, 1986:245-247), and late Oligocene (Arikareean) (P. montana Pierce, 1992), through the Quaternary. Pupoides costatus Pierce, new species (Fig. lO.A-lO.C) Diagnosis. — A small cylindrical Ischnopupoides, whorls short, sculptured with bold, regularly and widely spaced prosocline riblets, which are separated by two to five very fine costellae between each riblet on last whorls, and with a modestly everted, slightly thickened peristome. Description. — Shell dextral, of about 5.25 short whorls, very narrowly rimate, cylindric to slightly pupilliform with convex low conic apex, suture moderately impressed, and appearing crenulated by the riblets on later whorls. Nuclear whorls 1.5, smooth to finely punctate, neanic whorls initially with slender, regularly spaced prosocline costellae becoming more widely spaced riblets on later whorls, riblets spaced about six per mm on penultimate whorl and three to four per mm on body whorl, riblets separated by two to five (usually three) faint parallel costellae on last several whorls. Body whorl ascends moderately in last quarter whorl, slightly compressed laterally resulting in a weakly rounded keel, regaining a “U” shape at the prosoclinally oblique (—20°) aperture, peristome everted modestly, thickened within, thin callus on parietal margin, no tubercles or teeth. Dimensions of holotype: height 3.1 mm; diameter (estimated*) 1.5 mm; H/D 2.06; whorls 5.25. * — Shell laterally compressed, average of measurements with compression (1.9 mm) and normal to compression (1.1 mm). Discussion. — Both riblets and costellae appear slightly less prominent on spec- imens from the type locality than from the Canadian localities. There is little difficulty in placing Pupoides costatus in the subgenus Ischnopupoides. Pupoides costatus is quite similar to both early Oligocene Pupoides tephrodes Roth, 1986 from the Three Forks Basin of Central Montana and to late Oligocene Pupoides montana Pierce, 1992 from the Deer Lodge Basin of Southwestern Montana. In comparison, the costae of P. tephrodes differ in being quite constant in size, more closely spaced (12-14 per mm on last half of ultimate whorl), and consistently 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 29 much weaker and less prominent than the major riblets of P. costatus. The ap- erture of P. costatus ascends more strongly than that of P. tephrodes in the ulti- mate quarter whorl, and its peristome is more strongly everted and thickened within. Pupoides montana is larger, has a greater HAV ratio (ave. 2.29), and lacks riblets except on the last quarter whorl, where they occur directly behind a sharply terminated, unreflected peristome. Pupoides costatus differs from modern Pupo- ides hordaceus (Gabb, 1866) and P. inornatus Vanatta, 1915, in having shorter whorls, and in having widely spaced, very strong riblets separated by multiple fine costellae on later whorls. Etymology. — costatus, Latin — ribbed. Material. — More than 20 specimens. Holotype: CM-41634, and paratypes PPUP-139 (2 fragments) are from Wurtz Bend (NFF-19) locality, Flathead County, Montana, Kishenehn Formation, late Eo- cene-Oligocene. Additional paratypes: CM-41635 (3); KUMIP-289,714; KUMIP 290,715 (3); PPUP- 134; PPUP-135 (plastitype); PPUP-136 (3, 1 plastitype) are from Commerce Creek (CMC-1) locality; PPUP-137 (3 fragments) is from E-5 (NFF-1) locality; PPUP-138 is from E-4 (NFF-2) locality (1 fragment); and PPUP-174 from W-2 (NFF-5) locality, all British Columbia. All localities in Lower Member, Kishenehn Formation, late Eocene-Oligocene. Pupoides tephrodes Roth, 1986 (Fig. lO.D, lO.E) Description. — “Shell dextral, narrowly umbilicate, cylindric with convexly low-conic summit, about 5.2 tall whorls; suture moderately impressed, crenulated by axial riblets. Early whorls convex; fourth and fifth whorls roundly shouldered below suture, compressed at periphery and anteriorly. Nuclear whorls 1.5, smooth; neanic sculpture of slender, well-spaced, retractive riblets. Body whorl narrowly, slopingly shouldered, compressed at periphery, slightly attenuated toward base, rising gently on penult behind aperture. Aperture oblique, ovate, peristome narrowly everted, not thickened within; parietal margin oblique, covered by a thin callus. No angular tubercle present.” (Roth, 1986:247). Costae (riblets) evenly spaced on earlier whorls, becoming quite crowded, about 12-14 per mm, on last half of ultimate whorl. Measurements of three hypotypes from locality E-2 (NFF-6)(with average and range of dimensions): height, ave. 3.2 mm (3. 0-3. 4); diameter, ave. 1.42 mm (1.40-1.45); H/D, ave. 2.26 (2.14-2.43); whorls, ave. 5.2 (5.1-5.25). Discussion. — These specimens compare very well with the two paratypes of Pupoides tephrodes Roth, 1986 (USNM 377381) filed at the U. S. National Mu- seum. Examination of these paratypes showed that the costae are regularly spaced on the penultimate and earlier whorls as is observed on the specimens from the Kishenehn Basin, that they also shared the observed characteristic of crowding of costae on the last half of the ultimate whorl, and, although broken back and lacking complete apertures, one appeared to have a shallow constriction of the ultimate whorl immediately behind the aperture. Examination of a considerable number of Pleistocene and Recent P. hordaceus in the author’s reference collection showed that this crowding of the costellae is not a characteristic of P. hordaceus, and that the costae (riblets) of P. hordaceus are considerably stronger and somewhat more widely spaced than observed on the P. tephrodes specimens from either the Kishenehn or Three Forks basins. Material. — Eighteen specimens. Figured specimen CM-41632, and hypotypes CM-41633 (3); KU- MIP 289,712; KUMIP 289,713 (3); and PPUP-131 (7) are from E-2 (NFF-5) locality, British Colum- bia. Additional hypotypes are: PPUP-132 from Island (NFF-4) locality, British Columbia and PPUP- 133 (2) from Moose City South (NFF-10) locality, Flathead County, Montana. All localities in Lower Member, Kishenehn Formation, late Eocene-Oligocene. VOL. 70 Annals of Carnegie Museum Fig. 10.— STYLOMMATOPHORA: PUPILLIDAE. A-C: Pupoides costatus n. sp. A-B; Holotype (CM 41634), XI 2, laterally compresed; A. apertural; B. lateral; C. P. costatus Plastitype (PPUP-135), 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 31 Subfamily Gastrocoptinae Pilsbry, 1948 Genus Gastrocopta Wollaston, 1878 Type species. — Pupa acarus Benson, 1856, by subsequent designation (Pilsbry 1916-1918). Subgenus Gastrocopta s str. Gastrocopta minuscula Pierce, 1992 (in Pierce and Rasmussen, 1992:46-47) (Fig. lO.L, lO.M) Description. — “Shell minute, rimate, cylindric, apex obtusely conic, whorls less than five, moder- ately convex, sutures well impressed, sculpture of fine growth striae; body whorl rounded to faintly keeled, aperture suboval, slightly ascending, peristome thin, narrowly reflected, without ridge or crest behind aperture; ...” (Pierce and Rasmussen, 1992:46-47). Emendation to description. — Minor crest behind aperture, flattened externally over palatal folds, teeth five (2:1:2); angulo-parietal lamella bilobed, chevron shaped, lobe at mid-length deflected abax- ially, columellar lamella semi-discoidal, curved and descending to near vertical distally, palatal folds two, upper nodular, lower a deeply entering, elongate, erect, bluntly terminated semi-discoid. Range of measurements for hypotypes (N=14): height 1.5-1. 9 mm; width 0. 9-1.0 mm; H/W 1.58-2.00; whorls 4. 5-4. 8. Discussion. — Among modem species, G. minuscula closely resembles Carib- bean G. pellucida (Pfeiffer, 1841), differing in having only a small external dimple at the location of the lower palatal fold, essentially no crest, no basal fold, and a curved, descending columellar lamella. For further comparison, see Pierce and Rasmussen, 1992:47. Environment and Distribution. — This species was known only as a fossil from the Cabbage Patch fauna, late Oligocene-early Miocene, of southwest Montana. The overall climate and environment reconstmcted from the Cabbage Patch fauna is of a mixed shrubland (savanna) and woodland, seasonally wet and dry, with MAT of about 10°C, January mean probably not less than 5°C and July mean of about 15°C to 20°C and MAP near 50 cm. (Pierce, 1993:988-991). The modem analog, G. pellucida, in its various subspecies, is widely distributed in subtropical and tropical environments. It is found throughout the Antilles, Florida, with scat- tered colonies along the Atlantic Seaboard to Cape May, New Jersey, around the Gulf of Mexico generally at least to the Yucatan Peninsula (collected by senior author), across northern Mexico and southwestern United States to southern Cab ifornia, thence south throughout Baja California (Pilsbry, 1948:914; Smith et al., 1990:105-106). Modem G. pellucida hordeacella (Pilsbry, 1890) has been found in middle Pleistocene (Yarmouthian Interglacial) in North Texas (Pierce, unpub- lished data). X20 abapertural, detail of costae on plastitype; D, E: Pupoides tephrodes Roth, 1986, XIO, D. Hy- potype (CM 41632) apertural; E. Hypotype (KUMIP 289,712) apertural; F-H: Gastrocopta kintlaense n. sp. F-G; Holotype (CM 41640), XIO; F. apertural; G. lateral; H. G. kintlaense Paratype (PPUP- 151-2), X25, apertural detail; I-K: Gastrocopta akokala n. sp. I-J: Holotype (CM 41642) X15; I. apertural; J. lateral; K. G. akokala Paratype (PPUP-155a), XIO, excavated aperture; L-M: Gastrocopta miniscula Pierce, 1992, Hypotype (CM 41636), XI 5; L. apertural; M. lateral; N-O: Gastrocopta leonardi Pierce, 1992, Hypotype (CM 41638), X15; N. apertural; O. lateral, incomplete specimen; P- Q: Vertigo consteniusi n. sp. Holotype (CM 41644), XI 5; P. apertural; Q. lateral; R-S; Vertigo sp, c.f. V. arenula White, 1876, Hypotype (CM 41646), X15; R. apertural; S. lateral; T-U: Vertigo doliara n. sp. Holotype (CM 41648), X15; T. apertural; U. lateral; V-W: Vertigo micra n. sp. holotype (CM 41649), X15; V. apertural; W. lateral; VALLONIDAE. X-Z: Valionia kootenayorum n. sp. Holotype (CM 41651), X12; X. apical; Y. umbilical; Z. apertural XIO. 32 Annals of Carnegie Museum VOL. 70 Material. — More than 50 specimens. Figured specimen, CM-41636, and additional hypotypes: CM- 41637 (3); KUMIP 289,716; KUMIP 289,717 (4); and PPUP-140 (7) are from Wurtz Bend (NFF-19) locality, Flathead County, Montana. Additional hypotypes are as follows; PPUP=141 (17 steinkerns) from Kintla Creek Carnegie (KTL-1) locality; PPUP-142 from Lower Kintla Rapids (NFF-18) locality; PPUP-143 from Kintla Creek River (NFF-17) locality; and PPUP-147 (2) from Moose City South (NFF-10) locality, all Flathead County, Montana; and PPUP-144 from Commerce Creek (CMC-1) locality (1 mold and plastitype); PPUP-145 from E-5 (NFF-1) locality; and PPUP-146 (7) from E-2 (NFF-2) locality, all British Columbia. All localities in Lower Member, Kishenehn Formation, late Eocene-Oligocene. Subgenus Albinula Sterki, 1892 Type species. — Pupa contractu Say, 1822. Environment and Distribution. — Modem taxa of this subgenus are widely tol- erant in both temperature and moisture requirements although there appears to be a slight bias to grassland environments. It is widely distributed in North America east of the Continental Divide and has been reported from eastern Canada to north central Mexico. Known geologic range is from late Eocene-early Oligocene (Chadronian) (Roth, 1986:241), then from late Pliocene (Blancan) to Recent. A similar subgenus, Ameralbinula Pierce, 1992 was found in a late Oligocene-early Miocene (Arikareean) fauna (Pierce and Rasmussen, 1992:42). Gastrocopta kintlana Pierce, new species (Fig. lO.F-lO.H) Diagnosis. — A small ovate-conic Albinula distinguished by a fused, sinuous F- shaped, angulo-parietal lamella, a sub-vertically oriented discoidal columellar la- mella, and disparate palatal folds, upper small, tuberculate, lower massive, chisel- shaped in apertural aspect. Description. — Shell dextral, narrowly umbilicate, ovate-conic, of about 5.5 well rounded whorls, suture well impressed. Nuclear whorls about 1.6, finely punctate, juvenile whorls initially sculpted with faint striae that become stronger and closely and evenly spaced by the third whorl, continuing to the reflected peristome. Last quarter whorl ascending slightly, moderately but roundly keeled, in basal aspect nearly straight, slightly constricted behind peristome, weak crest. Aperture advanced about 20° past axis, roundly triangular, projecting, thickened by interior callus, upper insertion near hori- zontal, connected to lower insertion by diagonal callus, peristome reflected moderately at straight to slightly sinuous outer lip, inner lip reflected more strongly. Teeth five or, occasionally, six, angulo- parietal lamella fused, sinuous, L-shaped, extending to plane of aperture, proximal end nearly con- necting to peristome, projecting lobe near mid-length deflected adaxially, distal end curved abaxially, columellar lamella prominent, discoidal to roundly triangular, curves from sub-vertical to vertical distally, palatal folds two or three on callus, upper palatal small, nodular, shortly entering, lower palatal distinctively strong, chisel-like in apertural view, but tapering distally, more deeply entering than upper palatal, interpalatal not always present, thin, low, elongate ridge, descending diagonally to nearly join lower palatal. Dimensions of holotype: height 3.25 mm; diameter 1.8 mm; H/D 1.81; whorls 5.75; of paratype; height 2.8 mm; diameter 1.7 mm; H/D 1.65; whorls 5.25. Discussion. — This species blends many of the variations in apertural teeth noted in the subgenus Albinula, and is especially similar to modern Gastrocopta con- tractu V. climeana Vanatta, 1911, differing in being larger, in having less massive dentition, especially the angulo-parietal lamella, and a weaker callus connecting the insertions of the peristome. Among fossil Gastrocopta, G. kintlana closely resembles Pliocene G. proarmifera Leonard, 1946, differing in being considerably smaller, having weaker dentition, and an L-shaped angulo-parietal lamella. It also resembles, in outline, more elongate specimens of Oligocene G. obesa Pierce, 1992, but the dentition of the latter, subgenus Ameralbinula Pierce, 1992, is very different. Gastrocopta {Albinula) species a, Roth, 1986, is also quite similar, dif- 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 33 fering in size, in shape of sinuous angulo-parietal lamella, in having a horizontal columellar fold, and, apparently, lacking a palatal callus. Gastrocopta montana Roth, 1986 is much smaller, has a strong crest behind the aperture, and a more sinuous angulo-parietal lamella. Gastrocopta kintlana, and Roth’s G. montana and G. (Albinula) species a, are quite similar and, as Roth (1986:244) proposed, “G. contracta may be the surviving member of a once more diversified lineage.” Etymology. — Named for Kintla Creek, a tributary entering the Flathead River from Glacier National Park and flowing past the type locality. Material. — About 30 specimens. Holotype, CM-41640, and paratypes: CM-41641 (3); KUMIP 289,718; KUMIP 289,719 (3); and PPUP-148 (15, crushed and fragments) are from Kintla Creek- River (NFF-17) locality, Flathead County, Montana, Kishenehn Formation, middle Eocene-late Oli- gocene. Additional paratypes: PPUP-149 from North Ford Creek (NFF-21) locality; and PPUP-150 from Wurtz (NFF-19) Bend locality, both Flathead County, Montana; and PPUP-151 (fragments) from E-5 (NFF-1) locality and PPUP-152 (2 fragments) from Commerce Creek (CMC-1) locality, both British Columbia. All localities in Lower Member, Kishenehn Formation, late Eocene-Oligocene. Gastrocopta akokala Pierce, new species (Fig. lO.I-lO.K) Diagnosis. — A small ovate Albinula distinguished by a moderate to strong crest and a diagonal sulcus behind the aperture, a slightly bilobed, fused angulo-parietal lamella, and deeply entering lower palatal fold. Description. — Shell small, rimate, ovate with obtusely conic apex, of about 4.75 to 5.25 rounded, nearly smooth whorls, suture moderately impressed, last whorl about same diameter as penultimate. Nuclear whorls about 1.6, smooth to very finely punctate, subsequent whorls have occasional, very weak striae that become more regular and noticeable on the last whorl, base of last half whorl distinctly pinched or keeled, flaring to roundly triangular aperture with reflected peristome, moderate to strong crest behind aperture and distinct exterior diagonal sulcus behind crest marking position of palatal folds. Aperture with five or more teeth, angulo-parietal lamella arising near insertion of outer lip and only scarcely behind plane of the aperture, slightly bilobed anteriorly, sinuous, medial lobe deflected abaxially, inner continuation spiral or diverging slightly from columella, columellar lamella a small, elongate, blunt disk, near horizontal, palatal folds on slightly thickened diagonal ridge reflecting ex- terior sulcus, upper short, nodular, lower deeply entering, erect oval blade, ascending slightly distally, a tiny suprapalatal or infrapalatal node may occur. Dimensions of holotype: height 2.1 mm; width 1.3 mm; HAV 1.62; whorls 4.8. Range of measurements: height 1. 9-2.4 mm; width 1.1—1.45 mm; HAV 1.50-1.89; whorls 4.6-5.25. Discussion. — Gastrocopta akokala is intermediate in size and shape between modern G. holzingeri (Sterki, 1889 — -fide Pilsbry, 1948:883) and G. contacta (Say, 1822), and its apertural characteristics more closely resemble G. holzingeri, dif- fering in not having a strongly bifurcate angulo-parietal lamella. Among fossil specimens, Gastrocopta akokala appears most similar to the three albinulid spe- cies described by Roth (1986:241-244) from the late Eocene Bozeman Group of the Three Forks Basin of southwestern Montana. It differs from G. montana in being more ovate, lacking noticeable striae, having a stronger crest and deeper sulcus behind the aperture, and in shape and placement of the palatal folds; from G. sagittaria in being larger, ovate versus conical, having a smaller aperture, proportionally, and a strong crest and exterior sulcus; and from G. (Albinula) species a in being smaller, more ovate, with a strong crest behind the aperture, and an elongate lower palatal fold. Gastrocopta akokala superficially resembles, in shape, the considerably larger G. obesa Pierce, 1992, from the Oligo-Miocene Cabbage Patch Beds of the Renova Formation in the Deer Lodge and Flint Creek Basins of western Montana, but lacks the characteristic angulo-parietal lamella of the subgenus Ameralbinula Pierce, 1992. 34 Annals of Carnegie Museum VOL. 70 Etymology. — Named for Akokala Creek, a tributary entering the Flathead River from Glacier Na- tional Park just downstream of the Wurtz Bend type locality of this species. Material. — More than 100 specimens. Holotype, CM-41642, and paratypes: CM-41643 (10); KU- MIP 289,720; KUMIP 289,721 (10); PPUP-153, PPUP-154 (10), and PPUP-155 (16) are from type locality, Wurtz Bend (NFF-19) locality, Flathead County, Montana, Kishenehn Formation, middle Eocene-late Oligocene. Additional paratypes: PPUP-156 (2 crushed) from South Ford Creek (NFF- 22) locality; PPUP-157 (2 crushed) from North Ford Creek (NFF-21) locality; PPUP-158 (2 crushed) from Starvation Bend (NFF-12) locality; PPUP-159 from Bowman Creek (BWN-2) locality; PPUP- 160 (3 fragments) Kintla Creek Carnegie (KTL-1) locality; PPUP-161 (8) from Lower Kintla Rapids (NFF-18) locality; and PPUP-162 (2) from Moose City South (NFF-10) locality, all Flathead County, Montana; and PPUP-163 (4 plastitypes), PPUP-164 (4), PPUP-165 (4 plastitypes), and PPUP-166 (mold and plastitype) from Commerce Creek (CMC-1) locality; PPUP-167 (5) from E-5 (NFF-1) locality; PPUP-168 (>20) from E-4 (NFF-2) locality; PPUP-169 (2) from E-3 (NFE-3) locality (2); and PPUP-170 (6) from E-2 (NFF-6) locality, all British Columbia. All localities in Lower Member, Kishenehn Formation, late Eocene-Oligocene. Subgenus Vertigopsis, Cockerell mss., Sterki, 1892 Type species. — Pupa pentodon (Say, 1821). Gastrocopta sp., cf. G. leonardi Pierce, 1992, (in Pierce and Rasmussen, 1992:46) (Fig. lO.N, 10.0) Discussion. — Gastrocopta leonardi resembles modem G. pentodon but is larger and lacks a basal fold and crest. Among fossil species, it most resembles ‘'Ver- tigo’' praecoquis Russell, 1956^, from the Miocene of Wyoming, but differs in being a third larger, and in having a broader, more rounded aperture that ascends distinctly in the last 60° of the ultimate whorl. These specimens can be easily separated from the remaining pupillid taxa, but their fragmentary nature precludes precise identification. There are, however, enough similarities to G. leonardi spec- imens, from its type locality, to warrant this identification. Environment and Distribution. — This species is previously known only from the late Oligocene-early Miocene (Arikareean) Cabbage Patch fauna of southwest Montana. The overall climate and environment have been described under Gas- trocopta minuscula. The subgenus is distributed from southeastern Canada and eastern United States south to Florida, west to a center of diversity in Arizona and south along the Gulf and Caribbean coast of Mexico to Guatemala (Pilsbry, 1948:886-891). Geologic range of the subgenus includes the late Oligocene-early Miocene (Arikareean) and are common in middle to late Pliocene (Blancan) fau- nas. Referred material. — Four specimens. Figured hypotype is from CM-41638 (3), from Kintla Creek Carnegie (KTL-1) locality, Flathead County, Montana. Additional specimen CM-41639, from E-4 (NFF-2) locality, and one specimen, now lost, was from Commerce Creek (CMC-1) locality, both British Columbia. All localities in Lower Member, Kishenehn Formation, late Eocene-Oligocene. Subfamily Vertiginae Pilsbry, 1948 Genus Vertigo Muller, 1774 Type species. — Vertigo pusilla Muller, 1774. Subgenus Vertigo s. str. Environment and Distribution. — The subgenus Vertigo s. str. is widely distrib- uted from eastern and northern Canada to California and northern Mexico. It is a very tolerant group, being found from Sea Level to 3,000 meters (10,000 feet), from cool moist to hot semiarid climates. The geologic range of the subgenus is Eocene to Recent, with V. arenula known from the Eocene and Oligocene. The 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 35 subgenus is common in middle to late Pliocene (Blancan) sediments of the mid- continent and west. Vertigo consteniusi Pierce, new species (Fig. lO.P, lO.Q) Diagnosis. — A small, broadly ovate Vertigo with weak sculpture, roundly oval aperture with weak indentation on palatal margin, very long lower palatal fold. Description. — Shell small, rimate, broadly ovate, of about 4.6 to 5.2 well-rounded whorls, suture moderately impressed, juvenile whorls about 1.4, smooth to finely punctate, subsequent whorls with weak, fine striae; last half whorl roundly keeled, ascending slowly in last 30°, shallow, elongate sinus behind aperture marks location of palatal folds; aperture roundly oval, peristome thin and sharp, reflected over basal margin, very weak marginal sinus on palatal margin, weakly developed crest extending from sinus to keel, outer lip above sinus near parallel to axis, below sinus retracted about 15°; teeth four or more, parietal lamella thin, erect blade, arising near middle of parietal margin, slightly behind plane of aperture, columellar lamella near horizontal, blunt discoid, palatal folds two or more, upper elongate node, lower more deeply entering, long, straight, proximal third an erect, rounded lobe, distal two-thirds lower, thinner, near horizontal. Dimensions of holotype: height 1.80 mm; width 1.20 mm; HAV 1.50; whorls 4.6. Range of measurements, including holotype (N=9): height 1.7-2. 1 mm; width L1-L2 mm; HAV 1.50-1.75; whorls 4. 6-5. 2. Discussion. — Vertigo consteniusi appears to belong to the V. gouldi group {sen- su Pilsbry, 1948:971). The presence or absence of an angular lamella or basal fold could not be established. It differs from typical V. gouldi (A. Binney, 1843) primarily in strength of striae and length of lower palatal fold. Among fossil species, it resembles closely both late Oligocene V. whitei. Pierce, 1992, from which it differs in size, sculpture, and in lacking(?) an angular lamella; and Eocene to late Oligocene V. arenula (White, 1876), from which it differs in size, very blunt apex, and an even longer lower palatal fold and lack(?) of angular lamella. Both of the latter also appear to belong to the V. gouldi group, despite the presence of angular lamellae. Etymology. — consteniusi, so named to honor Norman and Lee Constenius, of Whitefish, Montana, dedicated paleontologists who supported in so many ways the field work for this study. Material. — Thirteen specimens. Holotype, CM-41644, and paratype KUMIP 289,722, are from Wurtz Bend (NFF-19) locality. Additional paratypes: PPUP-171 is from Bowman Creek (BWN-2) locality; CM-41645 (3), KUMIP 289,723 (3), and PPUP-172 (5) are from Kintla Creek Carnegie (KTL-1) locality, all localities Flathead County, Montana and in Lower Member, Kishenehn Formation, late Eocene-Oligocene. Vertigo sp., cf. V. arenula (White, 1876) (Fig. lO.R, lO.S) Discussion. — Unfortunately, the apertures of all specimens are damaged or sol- idly blocked, precluding precise identification; however the shape of the shell, the HAV ratio, the sculpture, and the sinus behind the aperture strongly suggest V. arenula. These specimens are a little larger than White’s (1876:46) description, 2 mm X 1.2 mm, (HAV 1.67) or that reported in the emended description by Pierce (Pierce and Rasmussen, 1992:47) of 2.14 mm X 1.25 mm, HAV 1.70, but closely approximate the measurements given by Yen (1946:500) of 2.5 mm X 1.5 mm (HAV 1.67). The close concordance of the H/W ratios is notable. Measurements of hypotype from E-3 (NFF-3) locality: height 2.50 mm; width 1.50 mm; HAV 1.67; whorls 5.3. It is possible that these specimens and those of Yen (1946:500) represent a distinct species, but the material at hand is inadequate to make this determination. 36 Annals of Carnegie Museum VOL. 70 Environment and Distribution. — Fossil from Eocene (White, 1876:105, 131) to late Oligocene (Pierce, 1993:fig. 2). This species is believed to be adapted to a climate with at least 10°C Mean Annual Temperature (MAT), January MAT prob- ably not less than 5°C, Mean Annual Precipitation 50 cm or less, and a grassland/ shrubland vegetation (Pierce, 1993:990). Material. — Seven specimens. Figured specimen: CM~41646 from E-3 (NFF-3) locality, British Co- lumbia. Additional hypotypes: CM-41647 (3); KUMIP 289,724 (3) from E-2 (NFF-6) locality, both localities in British Columbia, Lower Member, Kishenehn Eormation, late Eocene-Oligocene. Subgenus Indeterminate Vertigo doliara Pierce, new species (Fig. lO.T, lO.U) Diagnosis. — A small, extremely corpulent, oval Vertigo with a small, roundly triangular aperture. Description. — Shell small, obtusely oval with blunt apex and base, of about five rounded whorls, suture shallowly impressed, nuclear whorls about 1.5, smooth, subsequent whorls with very weak striae; body whorl markedly narrower than penultimate whorl, ascends slightly on penultimate in last 45°, roundly keeled with long, diagonal sulcus, extending across the keel, marking the position of the lower palatal fold; aperture small, roundly triangular, parietal lamella thin blade, arising near middle of parietal margin, deflected very slightly abaxially, neither angular, if it exists, nor columellar lamella, nor upper palatal fold observed, lower palatal fold long, deeply entering, sweeping abapically across base of whorl. Dimensions of holotype: height 2.05 mm; width 1.45 mm; HAV 1.41; whorls 5.0. Discussion. — This species probably belongs to the subgenus Vertigo s. str., but lacks adequate diagnostic features to so assign with complete confidence. In shape it resembles several corpulent species from the southeast United States, such as V. rugulosa Sterki, 1890, but none achieve the extreme corpulence, H/W 1.41, of this species, nor its distinct blunt oval shape. Among fossil pupillids, only among the Gastrocopta, e.g., G. sagittaria, Roth, 1986, H/W 1.38, and G. oviforma Pierce, 1992, H/W 1.48, is this degree of extreme corpulence achieved. Gastro- copta sagittaria has a much larger aperture, and G. oviforma is much larger. Etymology. — doliara, Latin, tubby, fat. Environment and Distribution. — Unknown. Material. — One specimen. Holotype: CM-41648 from Moose City South (NFF-10) locality, Flathead County, Montana, Kishenehn Formation, middle Eocene-late Oligocene. Vertigo micra Pierce, new species (Fig. lO.V, lO.W) Diagnosis. — A tiny ovate Vertigo with blunt apex and fine, distinct striae. Description. — Shell tiny, about 1.2 mm in height, rimate, ovate with bluntly conical apex, of about 4.7 well-rounded whorls, suture moderately impressed, sculpture of fine, distinct, retractive striae, last 60° of ultimate whorl rises on penultimate whorl, details of aperture and dentition unknown. Dimen- sions of holotype: height 1.25 mm; width 0.85 mm; HAV 1.47; whorls 4.7. Dimensions of plastitype: height 1.15; width 0.75, HAV 1.53. Discussion. — Vertigo micra is truly a pygmy among pygmies. The only other Vertigo of equal minuteness is modem V. hebardi Vanatta, 1912, which is a member of the subgenus Vertigo s. str., and of the Vertigo gouldi group. The type material at hand, three steinkems, all incomplete in the apertural area, and a cast, aperture not shown, are insufficient to determine precisely the taxonomic position 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 37 of Vertigo micra, but it appears highly likely that it is a member of the same subgenus and group. Among fossil Vertigo, V. micra is immediately distinguish- able on the basis of size alone; no other fossil Vertigo is so minute. However, in shape, it bears a strong resemblance to both Vertigo whitei Pierce, 1992, from the late Oligocene Deer Lodge Basin of far western Montana, and to Vertigo consteniusi, which is found in association with V. micra, but it is immediately distinguishable from both on size alone. Gastrocopta sagittaria Roth, 1986, from the late Eocene Three Forks Basin of western Montana, bears considerable re- semblance in shape of shell, but is, again, much larger. Etymology. — micra, derived from mikros, Greek, small. Environment and Distribution. — Unknown. Material. — Four specimens. Holotype, CM-41649, a near complete steinkern and paratype, CM- 41650 (2), are partial steinkerns from Kintla Creek Carnegie (KTL-1) locality, Flathead County, Mon- tana. Paratype PPUP-173, a mold, with multiple plastitypes, is from Commerce Creek (CMC-1) lo- cality, British Columbia. Both localities are in Lower Member, Kishenehn Formation, late Eocene- Oligocene. Family Valloniidae Pilsbry, 1900 Genus Vallonia Risso, 1826 Type species. — Helix pulchella Muller, 1774. Vallonia kootenayorum Pierce, new species (Fig. lO.X-lO.Z) Diagnosis. — Shell of the Vallonia cyclophorella Sterki, 1893 Group (sensu Pilsbry, 1948:1033), small, depressed, costate, aperture slightly descending, peri- stome weakly everted and thin, openly and eccentrically umbilicate. Description. — Shell thin, depressed, of about 3.125-3.375 whorls, nuclear whorls about 1.5, finely granulose, neanic whorls 0.5, very faintly, finely, and irregularly costate, adult whorls ornamented with thin, thread-like, costae, with yet finer costellae between each pair, major costae continue onto base only slightly diminished, about ten costellae in last millimeter behind aperture; aperture descend- ing slightly in last 15°, oblique, circular, peristome everted slightly, slightly thickened within; umbilicus wide, one-third major diameter, widening slightly in last one-half whorl. Measurements of holotype are: diameter, major, 2.05 mm; minimum, 1.75 mm; height 1.0 mm; HAV 0.49. Discussion. — Vallonia kootenayorum is clearly a member of the Vallonia cy- clophorella Group, ribbed, with a thin lip, and is most similar in size and sculpture to modern Vallonia perspectiva Sterki, 1893, from which it differs in having finer, more closely spaced costellae, an aperture that descends less, and it is consistently a little larger than average V. perspectiva. Taylor’s (1960:77) report of the four oldest occurrences of V. perspectiva ranged through much of the Blancan NAL- MA (from mid-Pliocene? to early Pleistocene, ca. 5. 0-1. 5 Ma). Vallonia kooten- ayorum is from late Eocene-Oligocene strata occurring in the northern Kishenehn Basin. The only other mid-Tertiary North American Vallonia is Vallonia berry i Pierce, 1992, from the late Oligocene-early Miocene Cabbage Patch beds of south- western Montana, 28-21 Ma. Vallonia berry i is a member of the very different V. pulchella (Muller, 1774) Group (Pilsbry, 1948:1023). The genus is known from Paleocene to Recent in Europe and Asia (Gerber, 1996:fig. 12). It is interesting to note that two of the three major North American valloniid groups (V. pulchella Group and V. cyclophorella Group) are represented in essentially modern form by the mid-Tertiary. Etymology. — Named to honor the aboriginal Kootenay Indians that lived in this area. 38 Annals of Carnegie Museum VOL. 70 Environment and Distribution. — The Vallonia cyclophorella Group originated in and is more generally distributed in the Mountain and Southwestern States and northern Mexico, but have spread spottily eastward throughout the Eastern United States (Pilsbry, 1948:1033-1037). Bequaert and Miller (1973:96-97; 191-192) describe V. perspectiva as a southwestern taxon, whereas V. cyclophorella is more montane in distribution. Vallonia perspectiva is characteristic of the Upper Son- oran and Transition Life Zones, whereas V. cyclophorella is more boreal, found in the Transition, Canadian and Hudsonian Life Zones. '"Vallonia perspectiva is a closely related sibling of V. cyclophorella, . . . both being clearly derived from a common ancestor” (Bequaert and Miller, 1973:96). Previously known geologic range of the V. cyclophorella Group is rather brief, from late Pliocene (Blancan) to Recent (Taylor, 1960:77). Material. — Thirty specimens. Holotype CM-41651 and paratype PVAL-110 are from Wurtz Bend (NFF-19) locality, Flathead County, Montana. Additional paratypes: PVAL-107 (7) from Commerce Creek (CMC-1) locality; PVAL-112, KUMIP 289,725 and KUMIP 289,726 (5) from E-4 (NFF-2) locality, both British Columbia; CM-41652 (9) and PVAL-108 from Kintla Creek Carnegie (KTL-1) locality; PVAL-109 (20 crushed) from Kintla Creek River (NFF-17) locality; and PVAL-111 from Bowman Creek (BWN-2) locality, all Flathead County, Montana, and all Lower Member, Kishenehn Formation, late Eocene-Oligocene. Superfamily Rhytidoidea Pilsbry, 1895 Family Haplotrematidae Baker, 1925 Genus Haplotrema Ancey, 1881 Type species. — Helix duranti Newcomb, 1 864. Haplotrema simplex (Russell, 1956), new combination (Fig. 9.G-9.H) Anguispira simplex Russell, 1956:111, fig. 5, plate 2, plate 3, fig. 1-3. Diagnosis. — A small Haplotrema with weak radial wrinkles and without spiral striae. Description. — Shell depressed, of about 5.5 broadly rounded, regularly expanding whorls, suture well impressed, umbilicus broadly open; surface glossy to waxy, light brown, embryonic whorls about 1.75, smooth, subsequent whorls with weak, low, irregular, discontinuous radial wrinkles, lacking spiral striae, aperture wider than tall, lip narrowly expanded, very slightly thickened. Measurements of hypotype (somewhat distorted): width 13 mm; height estimated 6 mm; HAV (est.) 0.46; whorls 5.5. Discussion. — Considering the fact that this genus is almost unknown as a fossil, its presence in this Oligocene age Kishenehn Fauna is exciting. The oldest fossil occurrence of this genus personally known are Wisconsinian age fossils from a terrace deposit from western Virginia. F. C. Baker (1920a: 389; 1920Z?:456) men- tions occurrences in Sangamonian age sediments in the Mid-continent. It is a huge jump in time back to the Oligocene, and remarkable to find specimens that so clearly belong to the living genus. Russell’s (1956:111, fig. 5, plate 2, fig. 1- 3) holotype (GSC 11610) is a peculiarly compressed and distorted specimen that is difficult to recognize as an Haplotrema. However, his two paratypes (GSC 11611 and 11612, loc. cit., Plate 2, figs. 4-6 and plate 3, fig. 1-3) are much more recognizable as Haplotrema, and either would have been more appropriate as holotype. The figured hypotype, also a topotype, is only slightly distorted and shows dorsal surfaces only, being imbedded in a matrix slab. It is too fragile to risk excavating the base of the specimen. The characteristic open umbilicus of this genus is obvious on other detached specimens. In Roth’s (1991) phylogenetic analysis of North American haplotrematids, 25 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 39 character states were analyzed, only four of which were of paleontologic value (Roth, 1991:163). His derived cladogram was used in substantiating a revision of the classification of the Haplotrematinae into two genera, Ancotrema H. B. Baker, 1930<3, and Haplotrema, with three subgenera under the latter, and in deriving an approximate age of 42.5 Ma (late middle Eocene) for the divergence of the “tem- perate Haplotrematinae and its tropical out-group, the Austroseliniti- nae, . . . ”(Roth» 1991:160). Based on the four shell characters used by Roth (1991:163), Haplotrema simplex fits neatly into the cladogram (Roth, 1991: fig. 3) as an (the?) ancestral species for the genus Haplotrema, with characteristics common to all three subgenera recognized, and sharing a sculpture pattern in common with H. minimum (Ancey, 1888), H. vancouverense (Lea, \^?>9-fide Pilsbry, 1946:221), H. keepi (Hemphill, 1890), and H. concavum (Say, 1821). It differs from each of the above in Character No. 25, not having “minute, close, wavy spiral striae” (Roth, 1991:163). Environment and Distribution. — The genus is widespread in North America with distinct eastern and western Groups (Pilsbry, 1946:204-225; Roth, 1991). It appears to be equally tolerant in environment, from cool, moist of Alaska and British Columbia to warm dry of California and Baja California (Pilsbry, 1946, loc. cit). A southern group, composed of the subgenus Haplotrema s. str., and H. caelatum have a pattern of distribution that is dominantly in the Coast Ranges from the San Francisco area into Baja California (Pilsbry, 1946:204-225). As a genus, it is known as fossil only from the late Pleistocene eastern United States (LaRocque, 1970:606). Material. — Fourteen specimens. Hypotypes (topotypes): CM-41653; KUMIP-289,727; and PHPL- 101 all from E-4 (NFF-2) locality, British Columbia. Additional hypotypes, CM-41654 (2); KUMIP- 289,728 (2); and PHPL-102 (2) all from E-5 (NFF=2) locality; and PHPL-103 from Commerce Creek (CMC-1) locality, all British Columbia, and PHPL-104 (4) from Starvation Bend (NFF-12) locality, Flathead County, Montana. Type horizon is Lower Member, Kishenehn Formation, late Eocene-Oli- gocene. Superfamily Bulimuloidea Tyron, 1867 Family Urocoptidae Pilsbry and Vanatta, 1898 Subfamily Eucalodiinae Crosse and Fischer, 1872 Genus Holospira von Martens, 1860 Type species. — Cylindrella goldfussi Menke, 1847, ICZN Opin., 1932. Subgenus Haplocion Pilsbry, 1902 Type species. — Holospira pasonis Dali, 1895a. Environment and Distribution. — The subgenus Haplocion is well adapted to existence in a warm, dry to arid environment. This group is distributed through southwest Texas and the northern Mexican states of Chihuahua, Coahuila, and Durango at moderate elevations to about 1,500 meters, with one species also reported from the Yucatan Peninsula (Bartsch, 1906:143; Pilsbry, 1953:151). Known geologic range of the subgenus is from Eocene {H. plumbea Roth and Megaw, 1989) to Recent, and, for the genus, from Cretaceous {H. dyeri Tozer, 1956) to Recent, with many species occurring during the Eocene and Oligocene. Holospira tabrumi Pierce, new species (Fig. II.A-ILE) Diagnosis. — A small Haplocion with prominent, well-spaced riblets on the body whorl, aperture a crudely pear-shaped oval with subparallel columellar and outer lips, peristome complete, strongly reflected, free of penultimate whorl. VOL. 70 Annals of Carnegie Museum 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 41 Description. — Apical whorls six, conic with rounded summit, embryonic whorls two + , smooth, rapidly expanding, next four whorls expanding regularly, well sculptured with fine but strong and regular prosocline costulae, subsequent four or more whorls cylindrical, diameter about 4.5 mm, penultimate and antepenultimate whorls about 1 .4 mm between sutures, bearing round-topped prosoc- line costae, body whorl 5.0 mm tall excluding reflected lip, bearing prominent, well-spaced, round- topped, slightly sinuous riblets spaced at about six riblets per millimeter on body whorl opposite aperture, umbilicus rimate to narrowly open, columella thin, minutely hollow; aperture projected and free of penultimate whorl, plane subparallel to axis, crudely oval pear-shaped with palatal and colu- mellar margins subparaliel, basal margin well rounded, merger of palatal and parietal margins roundly acute, peristome entire, widely reflected and strongly reinforced. Measurements: height (estimated) >15 mm; width of penultimate whorl, about 4.5 mm; height of aperture on palatal margin 4.2 mm, on columellar margin 3.4 mm; width 3.5 mm. Discussion. — Although no one specimen is complete, all have the essential characteristics of the subgenus Haplocion. Among modem species, the aperture and last whorls of H. tabrumi most resemble H. pasonis Dali, 1895, but is much smaller. Among fossil taxa, it is similar in profile to H. beardi n. sp., but H. beardi has finer, more closely spaced, round-topped costae on the body whorl, 10-12 per nun, and an aperture appressed to the penultimate whorl, the plane of which is retractive from the axis by about 25°, and a weakly reinforced and reflected intermpted peristome. Eocene H. plumbea Roth and Megaw, 1989, is very similar, but differs from H. tabrumi in being larger with a more fusiform outline, has weaker sculpture on the body whorl, an aperture whose plane deviates from the axis by 20°-35°, and peristome continuous although appressed to the penultimate whorl. It differs from Coelostemma dawsonae n. sp., and all other fossil Holospira, in being much smaller, in having weaker costae on penultimate and antepenultimate whorls, which become strong and obvious on the body whorl, and in having a roundly pear-shaped projected aperture. Etymology. — Named to honor Alan Tabrum, vertebrate paleontologist with the Carnegie Museum of Natural History, with whom I have worked in this project. Material. — More than 13 specimens, most incomplete or fragmentary. Holotype CM-41655, and paratypes CM-41656 (2); KUMIP 289,729: KUMIP 289,730 (2); PURC-101-106 are from W-2 (NFF- 5) locality, British Columbia, Lower Member, Kishenehn Formation, late Eocene-Oligocene. An apical fragment of seven whorls, PURC-107, from Moose City South (NFF-11) locality, Flathead County, Montana, same horizon, is provisionally referred to this species. Subgenus Haplocion{l) Pilsbry, 1902 Holospira beardi Pierce, new species (Fig. ll.F, ll.G) Diagnosis.— A small, conical-cylindric Holospira, aperture ovate, attached to penultimate whorl on parietal margin, peristome discontinuous, plane of aperture about 25° from axis. Description. — Shell small, cylindric-ovate with 12 or more whorls, embryonic whorls about two. Fig. 11.— STYLOMMATOPHORA; UROCOPTIDAE. A-E: Holospira tabrumi n. sp. A, B; Holotype (CM 41655), X5; A. apertural; B. lateral; C-E: Paratype (CM 41656 lot), X4; C. apical fragment; D. apertural fragment, frontal; E. apertural fragment, lateral; F-G: Holospira beardi n. sp. Holotype (CM 41657), X5; F lateral; G. apertural; H-K: Coelostemma dawsonae n. sp. X3; H-I: Paratypes, recon- struction composed of PURC-108, apical fragment, dextral, and KUMIP 289731, apertural fragment, dextral; J: Paratype KUMIP 289732, sinistral; K: Holotype CM 41658, dextral. POLYGYRIDAE. L. Ashmunellal hilli n. sp. Holotype (CM 41671), X5, apertural fragment with parietal lamella; M-O: Praticolella lucifera n. sp. Holotype (CM 41669), X5; M. apical; N. umbilical; O. apertural. 42 Annals of Carnegie Museum VOL. 70 smooth, next four whorls with fine prosocline costellae becoming less pronounced on cylindrical whorls, then more pronounced on last two whorls with about 10-12 costae per millimeter on body whorl. Apical six whorls rapidly expanding, beehive-shaped, later whorls near cylindrical, exposed height of antepenultimate and penultimate whorls about 1.9 mm, body whorl about 5 mm in height, slightly constricted behind aperture, roundly keeled, umbilicus rimate. Aperture “U” shaped, peristome broadly reflected and strongly reinforced, but not continuous over penultimate whorl between inser- tions, parietal insertion much above columellar insertion, meeting penultimate whorl at about 60° near mid-point, bending rapidly to parallel axial plane, basal lip well rounded, plane of aperture retractive at about 25° from axis of shell, no lamellae detected; columella shallowly spiral and minutely hollow. Dimensions of holotype: height (including expanded lip), 15.25 mm; width (specimen slightly crushed) of penultimate whorl, estimated, 4.0 mm; exterior height of aperture at parietal insertion, 4.8 mm; at columellar insertion, 3.5 mm; width (slightly deformed), 2.9 mm. Discussion. — The embryonic whorls of the essentially complete holotype are slightly damaged, but are definitely not mammillate. This species seems to blend the characteristics of several genera and subgenera from among the Urocoptidae. The slender, minutely hollow columella, along with the retained spire and smooth embryonic whorls argues for placement in the Holospirinae and the genus Ho~ lospira, whereas the incomplete peristome suggests relationship to the Microcer- aminae Pilsbry, 1904. Among modem species of Holospira, the most similar are of the subgenus Haplocion Pilsbry, 1902, as later redefined (Pilsbry, 1953:151), or Allocoryphe (Pilsbry, 1946). Holospira pasonis Dali, 1895, type species for Haplocion, differs from H. beardi in being larger, having a more convex profile, and a peristome complete and free of the penultimate whorl. Holospira minina von Martins, 1897, type species fox Allocoryphe, is smaller, strongly ribbed and also has a peristome complete and free of the penultimate whorl. Among fossil taxa, H. beardi apparently coexisted with and most resembles H. tabrumi, but H. tabrumi differs in bearing coarser costae, six per mm on body whorl opposite the aperture, and an aperture with complete peristome free of the penultimate whorl, the plane of which is subparallel to the axis. Coelostemma dawsonae n. sp., found at nearby localities, is larger, has fewer whorls, and a very different aperture. Cretaceous H. dyeri Tozer, 1956 is much larger and differs in nearly all aspects. Eocene H. grangeri Cockerell, 1914 appears about the same size, but has a more fusiform outline, and is too incomplete to merit further comparison. Oligocene H. eva Gardner, 1945, is similar, but larger, and has a continuous peristome that is adnate to the penultimate whorl. The enigmatic Pupal leidyi is also about the same size, but lacked the aperture and has now been lost. White’s (1886:27, plate 5, figures 8-10) Pupal leidyi Meek, 1873, which may or may not be the same, is similarly convex, and has a projected peristome free of the penultimate whorl. La Rocque’s (1960:49) Paleocene spec- imens of Holospira cf. H. leidyi are too imperfect for other than tentative assign- ment to genus. Eocene H. plumbea Roth and Megaw, 1989, is very similar, but differs from H. beardi in being larger with fewer whorls, having a more fusiform outline, and a peristome continuous although appressed to the penultimate whorl. In apertural character, H. plumbea seems to occupy an intermediate position be- tween the primitive(?) aperture of H. beardi, and the produced aperture with peristome complete and free of the penultimate whorl of H. tabrumi and typical of the modern subgenus Haplocion. Etymology. — Named to honor Chris Beard, Curator of Vertebrate Paleontology at the Carnegie Museum of Natural History, with whom I have shared work and laughs on this project. Material. — One specimen. Holotype: CM-41657, W-2 (NFF-5) locality, British Columbia, Lower Member, Kishenehn Formation, late Eocene-Oligocene. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 43 Genus Coelostemma Dali, 1896 Type species. — Holospira elizabethae Pilsbry, 1889. Subgenus Coelostemma s. str. Coelostemma dawsonae Pierce, new species (Fig. ll.H-ll.K) Diagnosis. — A large, squat, ambidextrous Coelostemma with costulae/costae on all whorls after smooth, rounded embryonic whorls. Description. — Shell large, squat, cylindrical to slightly fusiform with roundly conical apex, ambi- dextrous, both dextral and sinistral morphs found in association; embryonic whorls 2.0 to 2.2, rapidly expanding, smooth, somewhat protruding and rounded, subsequent four apical whorls expanding with fine, closely spaced, sweeping prosocline costulate, becoming more widely spaced, sharp-topped costae on cylindrical body whorls, with about five costae per mm on body whorl, body whorl about 5 mm tall, penultimate and antepenultimate whorls each about 2.5 mm between sutures; umbilicus narrowly open, columella large, hollow, almost Vs diameter of shell; body whorl with rounded base, straightening and projecting forward, slightly constricted laterally behind roundly pear-shaped aperture, aperture about 7 mm tall and 6 mm wide, peristome complete, free of penultimate whorl, well reinforced and reflected on palatal and basal margins, less on parietal and columellar margins, no apertural lamella observed. Measurements of dextral holotype (crushed): height (estimated) >17.5 mm; width (at eighth whorl) 8 mm; whorls 10.5. Measurements of sinistral paratype (crushed): height (estimated) >19.0 mm; width (at eighth whorl) 7.5 mm; whorls >11. Discussion. — Among living taxa, C. dawsonae most resembles the group of southwestern Mexican Coelostemma, differing in presenting a squatter profile as a result of fewer whorls and a slightly greater diameter of the cylindric body. Among fossil taxa, the general size matches Holospira adventica Russell, 1956, which he found at two nearby Canadian localities (Russell, 1956:110), but both localities are now either destroyed or covered by slumping. Holospira adventica differs greatly from C. dawsonae in apertural character, in that its aperture is appressed to the penultimate whorl, peristome incomplete and/or interrupted by a prominent superior (parietal?) lamella, and it has an unusually tall body whorl, characters not characteristic of holospirids. Holospira grangeri Cockerell, 1914 is smaller and more slender. Holospira eva Gardner, 1945 is much taller and has a more elongate profile. Pilsbry (1953:139-140) determined that H. eva belongs in Holospira s. str. Holospira dyeri Tozer, 1956 is much taller, has a decidedly bullet-shaped profile, and differs in almost all characters. The enigmatic Pupal leidyi Meek, 1873, although never figured, is, by the original description, a shorter, narrower shell (14 mm X 5 mm), with about 15 whorls (Meek, 1873:517). Etymology. — Named to honor Mary Dawson, Curator of Vertebrate Paleontology at the Carnegie Museum of Natural History, with whom I have shared many an outcrop in seeing this project to a conclusion. Environment and Distribution. — The genus Coelostemma is well adapted to existence in dry to arid environments. Its geographical distribution is unusual, one group is found in the high arid areas of the states of Coahuila and Nuevo Leon in northeastern Mexico, whereas a second group, isolated from the first by some 750 km, is found in similar terrain in the states of Guerrero and Oaxaca of southwestern Mexico. Coelostemma dawsonae most resembles the latter, and more distant, group. Geologic range of the genus is uncertain. Roth and Megaw (1989: 7) suggested that the enigmatic Pupal leidyi may be a Coelostemma. If so, the geologic range of the genus would have included the Eocene. Material. — More than 20 specimens. Holotype, dextral: CM-41658; paratypes, dextral: KUMIP- 289731; PURC-108; PURC-110 (7 apices); PURC-111 (3 fragments); and PURC-114(8); and para- 44 Annals of Carnegie Museum VOL. 70 types, sinistral: CM-41659; KUMIP-289732, PURC-109, are from Starvation Bend (NFF-12) locality, Flathead County, Montana, Lower Member, Kishenehn Formation, late Eocene-Oligocene. Additional specimens, apical fragments of three or fewer whorls, and probably assignable to this taxon, are as follows; PURC-112, from E-3 (NEE-3) locality; and PURC-113 (4) from E-4 (NFF-2) locality, both in British Columbia, same horizon. Superfamily Arionoidea Gray, 1840 Family Discidae Thiele, 1931 Genus Discus Fitzinger, 1883 Type species. — Helix ruderata H. and A. Adams, 1857. Discus mackenziei Russell, 1956fl (Fig. 9.M) Discussion. — Russell (1956^:111-112, figure 6, plate 3, figs. 4-6) found this species rather rare and only at his Locality E-4. His restoration in his Figure 6 appears generally accurate, although somewhat more depressed and flattened than justified by either his type material or the specimens collected on this project. Russell {loc. cit.) compared D. mackenziei to living D. patulus (Deshayes, 1830). Without a bit of care and caution, D. mackenziei can be easily confused with its fellow discid, Anguispira, cf. A. alternata, which is found at other localities in this area. The easiest way to separate incomplete or immature shells of the taxa is in the embryonic and neanic whorls. For Anguispira, cf. A. alternata, the nu- cleus is rather strongly produced above the subsequent whorl, the nuclear diameter is 0.5 mm, and the diameter of an immature or broken back specimen at 3.5 whorls will be 6.5 mm, or more. For D. mackenziei, the nucleus is scarcely elevated above the subsequent whorl, the nuclear diameter is less than 0.4 mm, and the diameter at 3.5 whorls is about 4.5 mm. This size difference is propor- tional throughout. At any given whorl, the whorl diameter for D. mackenziei will be noticeably smaller than at the same whorl locality for Anguispira, cf. A. al~ ternata, and the costulae of D. mackenziei are finer and more closely spaced. Environment and Distribution. — The modern genus is Holarctic, widely spread across the northern States and Canada, with relict populations extending south in the mountains of the East and Southwest and, in general, prefers cooler, moist climates and wooded environments (Hubricht, 1985:19, 20, 106-109; Pilsbry, 1948:598-622). Extant species are known from the mid-Pliocene of Wyoming and the late Pliocene of Nebraska (both Blancan). Fossil species are known to range from upper Cretaceous (as Gonyodiscus Fitzinger, 1833) through Miocene (Pilsbry, 1948:599) and Recent. Material. — Two specimens and fragments. Figured specimen, KUMIP-289,733 (1 Tfragments), and additional hypotypes: CM=41660 (1); and PDIS-101 (fragments) are from E-4 (NFF-2) locality, Brit- ish Columbia. Additional hypotype, PDIS-102 (fragments) from Kintla Creek-Carnegie (KTL-1) lo- cality, Flathead County, Montana. Both locations are in the Lower Member of the Kishenehn For- mation, late Eocene-Oligocene. Genus Anguispira Morse, 1864 Type species. — Helix alternata (Say, 1816). Anguispira sp., cf. A. alternata (Say, 1816) (Fig. 9.N-9.P) Description. — Shell depressed heliciform, sub-lenticular, openly umbilicate, about Vs of diameter, with six or more rounded whorls; embryonic whorls about 1.5, first whorl smooth to faintly cross- hatched, with weak prosocline wrinkles on last half whorl, neanic whorls with distinct, regular, riblets. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 45 somewhat carinate, becoming more rounded on later whorls; aperture unknown, shell material bleached, no evidence of flame-like blotches. Measurements of hypotype: height ±8.5 mm; width >16 mm, estimated 18 mm; HAV (est.) 0.47; whorls >6. Discussion. — The largest specimens are usually quite abraded and the riblets are not well detailed. Smaller, juvenile or broken shells have embryonic whorls and riblet patterns that are not distinguishable from the modern species (Fig. 9. 0,P; KUMIP 289,734). Despite the lack of a complete shell, or of a usable ap- erture, there appears essentially no doubt that, conchologically, these specimens are within the normal range of variability of Recent A. alternata {s. lat.), and is therein so placed, albeit reluctantly. F. C. Baker (1920:389) reports this species from strata as old as “Aftonian”, but the oldest occurrence of the species with which the author is personally aware is the University of Nebraska State Museum locality NO- 104, Nuckolls, County, Nebraska, dated as late Kansan (ca. 600 Ka) on the basis of the immediately underlying Pearlette Type O ash. In either event, there is a huge time gap between these Kishenehn specimens and the Pleistocene occurrences. The distributional anomaly is just as great. Recent A. alternata is definitely an eastern taxon, not being found west of about 97° West longitude, but is distributed over a wide range of latitude and climate zones, from eastern Canada to coastal Texas (Pilsbry, 1948:569). A. holroydensis Russell, 1956^1 from the Upper Miocene North Park Formation of southeastern Wyoming has fewer whorls, four, is smaller, diameter 7.9 mm, and has subangular periphery, all characteristics of an immature or broken back specimen. Russell commented that, “Shells very similar to the present ones in appearance are found in the living Anguispira alternata (Say) and A. cumberlan- diana (Lea)” (Russell, 1956n:1262). Paleocene Anguispira russelli Tozer, 1956, from photographs and description (Tozer, 1956:84-85, plate 8, figs. lOa-lOc), also appears to fall within the range of variation of the Recent species, and differs from these Oligocene specimens only in having somewhat fuller, more rounded whorls. Not having seen the type material for either of these species, I do not include them in synonymy, although I strongly suspect they so belong. In any event, a species of land snail that has apparently resisted morphological change for 60 Ma would certainly remarkable. Environment and Distribution. — The living species is widely distributed in east- ern North America, generally east of the 100th meridian and north of the 35th parallel into southern Canada (Hubricht, 1985:104). “A prolific, ubiquitous, and hardy snail favoring woodlands, especially of deciduous trees, but able to live in drier, more open woods or fields” (LaRoque, 1970:671). The geologic range of the genus is not clear. Species appearing to be essentially identical conchologically are found in Paleocene (A. russelli) and Miocene (A. holroydensis). The species is previously known as fossil from middle Pleistocene to Recent. Material. — Nine specimens. Figured specimens: CM-41661, a large, worn specimen, is from Com- merce Creek (CMC-1) locality, British Columbia, and KUMIP-289734 is from Wurtz Bend (NFF-19) locality, Flathead County, Montana, both Kishenehn Formation. Additional specimens: CM-41662, and PDIS-104 are from Starvation Bend (NFF-12) locality; KUMIP-289735, and PDIS-103 (2 frag- ments) are from Wurtz Bend (NFF-19) locality; and PDIS 105 is from South Ford Creek (NFF-22) locality, all Flathead County, Montana, and all are Lower Member, Kishenehn Formation, late Eocene- Oligocene. 46 Annals of Carnegie Museum VOL. 70 Superfamily Limacoidea Family Limacidae Gray, 1824 Genus Deroceras Rafinesque, 1820 Type species. — Limax laevis Muller, 1774. Deroceras mahiz Pierce, 1992 (Fig. 9.K, 9.L) Discussion. — The specimens from the Kintla Creek-River locality are essen- tially indistinguishable from the type material from the lower Cabbage Patch Beds (ca. 28 Ma) of the Deer Lodge Basin of southwestern Montana. These specimens generally retain at least part of their chalky outer layer, suggesting little or no transport. Larger plates (<2.5-3 mm) from most of the other localities tend to have a somewhat more rounded and less projecting base than observed in the Cabbage Patch type material or in the specimens from the Kintla Creek-River locality, and their basal yellow crystalline calcitic layer is somewhat thinner. Such minor variation is not thought to have taxonomic significance. These specimens somewhat resemble D. aenigma Leonard, 1950, but D. mahiz can be easily dis- tinguished by its greater size, in having the nucleus more centrally located at the posterior margin, and in lacking the notch that is found beneath its nucleus. Small (~1 mm), circular, yellow calcitic disks, with or without a chalky dorsal surface, becoming elongate (1mm X <2mm) with growth, found at several localities are believed to be embryonic plates of this species. Environment and Distribution. — Genus is of Holarctic distribution plus South America (LaRocque, 1970:664). Pilsbry’s (1949:539) Deroceras laeve (Muller, 1774) Group is a plausible modern analog and probable descendant. Species of this Group are distributed from Alaska and Baffin Island to Central America (Pilsbry, 1949:539-557), and prefer humid areas like floodplains and marshes but can survive in drier conditions with some protection from desiccation (LaRocque, 1970:667). This species was previously known only from the late Oligocene Cab- bage Patch Fauna (late Arikareean) of southwestern Montana (Pierce, 1992). As a genus, it is well represented from Oligocene to Recent. Material. — More than 40 specimens. Figured specimen: CM-41665, and additional specimens CM- 41666 (5), and KUMIP-289738 (5), are from Kintla Creek-River (NFF-17) locality. Additional spec- imens: PLIM-114 (17) from Commerce Creek (CMC-1) locality; PLIM-115 (4) from E-4 (NFF-2) locality; PLIM-116 from W-2 (NFF-5) locality, all British Columbia; KUMIP-289737 and PLIM-117 (7) from Starvation Bend (NFF-12) locality; PLIM-118 from Wurtz Bend (NFF-19) locality; PLIM- 119 (4) from Lower Kintla Rapids (NFF-18) locality; PILM-120 from Bowman Creek (BWN-2) locality, all Flathead County, Montana. All localities in Lower Member, Kishenehn Formation, late Eocene-Oligocene. Most of the PLIM specimens are very small, embryonic to immature. Deroceras securis Pierce, 1992 (Fig. 9.1, 9.J) Discussion. — As was the case with Deroceras mahiz, the specimens of D. se- curis are essentially indistinguishable from those of the type locality in the middle Cabbage Patch beds (ca. 24 Ma) of the Flint Creek Basin of southwestern Mon- tana. Most specimens are somewhat abraded, the outer chalky layer has been removed, and the growth wrinkles are smoothed, suggesting transport of some distance. Environment and Distribution. — As with Deroceras mahiz. Material. — Twenty-five specimens. Figured specimen: CM-41663, and additional specimens CM- 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 47 41664 (7), KUMIP-289,736 (7), and PLIM-113 (10), all from Kintla Creek-Camegie (KTL-1) locality, Flathead County, Montana, Lower Member, Kishenehn Formation, late Eocene-Oligocene. Superfamily Zonitoidea Morch, 1864 Family Zonitidae Morch, 1864 Subfamily Zonitinae Morch, 1864 Genus Nesovitrea Cooke, 1921 Type species. -—Vitrea pauxilla Gould, 1852. Subgenus Perpolita H. B. Baker, 1928 Type species. — Nesovitrea hammonis (Strom, 1767). Nesovitrea pulchra Pierce, 1992 (Fig. 9.Q, 9.R) Emended description. — Shell strongly depressed, of nearly four regularly increasing whorls, suture scarcely impressed. Embryonic whorls about 1.5, smooth, subsequent whorls with apical sculpture of regular, weak, prosocline growth wrinkles, on some specimens more widely spaced, fine prosocline radial striae appear at about the third whorl, these striae gradually become more prominent and more widely separated, up to 0.5 mm apart just behind aperture, last whorl strongly convex below, base with slightly weaker wrinkle striae. Aperture deeply and obliquely lunate, outer lip retractive about 30° from vertical axis, peristome simple, acute, umbilicus open, initially narrow conical, enlarging with last whorl to become funnel-shaped, contained four to five times in diameter. Measurements of figured hypotype, CMNH 41667: width, major 3.6 mm, minor 2.7 mm; height (estimated, slightly crushed) 1.8 mm; HTW (est.), 0.50; whorls 3.5. Discussion. — The type specimens from the Cabbage Patch molluscan local fau- na of southwest Montana, although plentiful and in good condition, are entirely replaced and recry stalized. The specimens herein from the Kishenehn Basin, al- though crushed and variously compressed, are composed of original shell material. The finer details of sculpture are revealed, resulting in the emended description, above. As previously concluded, N. pulchra appears to be closely related and possibly ancestral to the modem palearctic N. hammonis (Strom, 1767) group. Environment and Distribution. — Palearctic N. hammonis is a plausible modern analog of N. pulchra. In Kansas, it “is an inhabitant of woodlands where it lives in decaying leaves, beneath loosened bark on dead trees and under sticks and fallen logs, . . . where the annual rainfall is generally more than 35 inches, but it declines in frequency toward the more arid Plains Border, . . . even where timber is locally available” (Leonard, 1950:37). In North America, modem subspecies of this group are found as far south as the 37th parallel (Virginia to Colorado), with probably relic populations in New Mexico and Arizona (Bequaert and Miller, 1973:67, 68, 145). As fossils, N. hammonis is known from late Pliocene (Blancan) to Recent faunas (Pierce, 1992:612). Nesovitrea pulchra was known previously only from the late Oligocene (early Arikareean) Cabbage Patch fauna of south- western Montana (Pierce, 1992). Material. — Forty five specimens. Figured hypotype, CM-41667 is from E-4 (NFF-2) locality, British Columbia. Additional specimens; PZON-104 (2), PZON-105 (14) Commerce Creek (CMC-1) locality; PZON-106 from E-5 (NFF-1) locality; CM-41668 (5) from E-4 (NFF-2) locality, all British Columbia, and KUMIP-289,739, KUMIP-289,740 (6) from Wurtz Bend (NFF-19) locality, Flathead County, Montana. All localities in Lower Member, Kishenehn Formation, late Eocene-Oligocene. Regretably, several specimens from Kintla Creek-River locality that were examined and listed are now missing. This locality is retained on the Kishenehn Fauna List, Table 2. 48 Annals of Carnegie Museum VOL. 70 Superfamily Polygyroidea Pilsbry, 1895 Family Polygyridae Pilsbry, 1895 Subfamily Polygyrinae Pilsbry, 1895 Genus Praticolella von Martens, 1892 Type species. — Praticolella ampla (Pfeiffer, 1841). Praticolella lucifera Pierce, new species (Fig. ll.M-ll.O) Diagnosis. — Shell small, few whorled, low domed for genus, peristome lightly reinforced and boldly expanded. Description. — Shell small, glossy pale yellowish brown, low domed depressed helicoid with about 4.2 rounded, regularly expanding whorls, suture moderately impressed, embryonic whorls about 1.5, smooth to very finely punctate, subsequent whorl with fine, regular, slightly prosocline costellae, about 20 per mm, becoming faint and indistinct on remaining whorls; last 0. 1 to 0.2 whorl expanding more rapidly, not descending, with four or five faint spiral striae on base of whorl not observed on preceding whorls, aperture ovate lunate, peristome boldly reflected, slightly thickened within, light callus on parietal margin, reflected aperture partially blocking narrow umbilicus. Measurements of holotype: width, major 7.2 mm; width, minor, est. 6.8 mm; height, est. 3. 5-4.0 mm, crushed; whorls, 4.2. Discussion. — Praticolella lucifera is remarkably similar to modem P. griseola (Pfeiffer, 1841), especially as found on the Mexican Yucatan Peninsula. Prati- colella lucifera differs in being slightly smaller, having fewer whorls, 4.2 versus 5.0 for P. griseola, in having bolder and more regular sculpture on the neanic whorls, and in the more strongly and widely reflected peristome. It is easily sep- arated from P. berlandierana (Moricand, 1833) by its greater size, more whorls, and strongly reinforced peristome, and from P. campi by its smaller size and weakly reflected peristome. In general, P. lucifera shows close affinities to the Section Praticolella {s. str.) group {sensu Thiele, 1992:932), found from Texas to Nicaragua, but is quite distinct from the other section of this species that occurs east of the Mississippi River. Praticolella prisca Auffenberg and Portell, 1990, known from the middle Miocene of Florida, is the only other pre-Tertiary occur- ence of this genus known. Environment and Distribution. — Praticolella griseola, the closest modem an- alog, prefers warm, open, or lightly wooded, grassy zones and, although preferring relatively humid conditions, is tolerant to seasonal drought (Pilsbry, 1940:690- 695). It is generally distributed from South Texas through eastern Mexico to Nicaragua. Cheatum and Fullington (1971:37) record P. berlandieriana as a Pleis- tocene fossil. Etymology. — lucifera, Latin, shiny. Material. — Two specimens. Holotype: CM 41669, and CM 41670, a fragment, are from Starvation Bend (NFF-12) locality, Flathead County, Montana, Lower Member, Kishenehn Formation, late Eo- cene-Oligocene. Subfamily Triodopsinae Pilsbry, 1940 Genus Ashmunella Pilsbry and Cockerell, 1899 Type species. — Ashmunella rhyssa miorhyssa (Dali, 1898). Ashmunella{l) sp. (Fig. ILL) Description. — Shell a slightly depressed helicoid, moderately large for genus, with about five rapidly expanding, rounded whorls, suture moderately impressed, openly umbilicate, ultimate whorl bearing regular, collabral, smoothly rounded ribs spaced at about 0.5 mm intervals, ribs more prominent and stronger on base of whorl; aperture descending moderately in last 3 mm, outer lip well reinforced. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 49 narrowly expanding, rather bluntly terminated, incomplete except for insertion into penultimate whorl which bears a thin callus across parietal region, and a large (—4.5 mm long by 2 mm high), diagonally situated parietal lamella, located well in front of basal lip and partly blocking aperture, with the shallowly sigmoidal apex of the lamella deflected slightly into aperture, outer lip oblique, at 30° angle from the axis. Measurements of holotype (broken and crushed) are as follows: diameter, —15 mm; height —8.0 mm; HAV —0.55; whorls about 5. Discussion. — Reconstruction is based on a basal fragment which includes more than one-half of the ultimate whorl, with part of the collapsed whorls of the spire in cross section, and a small portion of the inner lip of the aperture, the parietal lamella, and the umbilicus. Had not the parietal lamella been observed, the su- perficial resemblance of this specimen to Helminthoglypta Ancey, 1887 and the known presence of this genus in nearby fossil localities of similar age might have resulted in tentatively placing this species in that genus. However, a combination of factors, i.e., size, shape, ornamentation, aperture, and parietal lamella, surely places this species in the Polygyridae, and, most likely, in the Triodopsinae. The shells of the genera of the Triodopsinae are rather similar, and the placement of this specimen in a genus is somewhat questionable. This specimen seems to com- bine conchological traits of the modem genera Ashmunella, Vespericola Pilsbry, 1939, and the subgenus Cryptomastix Pilsbry, 1939 of Triodopsis Rafinesque, 1819. All of these share many conchological characters in common, but each can be distinguished by unique characters that are often difficult to distinguish. This species has the strong, robust parietal lamella characteristic of Cryptomastix, the somewhat heavier shell of Vespericola, and the open umbilicus of Ashmunella, and the strong sculpturing of the ultimate whorl not noted among the Cryptomastix or the Vespericola. In fact, this species would make a good progenitor for the western polygyrids. Based on the openly umbilicate nature of this specimen, which is probably a primitive character, the strong sculpture of the ultimate whorl, and the interpreted environment in which encountered, it appears more closely related to the genus Ashmunella, and is so assigned herein, but with question. Among extant members of the genus, the sculpture of Ashmunella{l) sp. resembles that encountered among the subspecies of A. townsendi Bartsch, 1904. Environment and Distribution. — Modem distribution is limited to far West Tex- as, southwestern New Mexico, southeastern Arizona, and northern Chihuahua. “Most species live at moderate elevations (4,000-6,000 feet), but a few reach 9,000-12,000 feet (e.g. in New Mexico). They prefer well- sheltered and shaded biotopes in sloping talus of loose soil and rock debris” (Bequaert and Miller, 1973:37). Fossil occurrences of the genus prior to the Pleistocene have not been reported (Bequaert and Miller, 1973:40). Material. — One specimen. Illustrated type CM 41671 is from Commerce Creek (CMC-1) locality, British Columbia, Lower Member, Kishenehn Formation, late Eocene-Oligocene. Superfamily Camenoidea Pilsbry, 1895 Family Oreohelicidae Pilsbry, 1939 Genus Oreohelix Pilsbry, 1903 Type species. — Helix strigosa Gould, 1846. Discussion. — The generally accepted conchological differences between the modern genera Oreohelix and Radiocentrum Pilsbry, 1905, have been defined by Pilsbry (1939:418) as follows: “Embryonic shell of one and one half radially costulate whorls . . . Subgenus Radiocentrum’, Embryonic shell of more than two whorls at birth, variously wrinkled and striate . . . Subgenus Oreohelix s. str.'' On 50 Annals of Carnegie Museum VOL. 70 the same page, in his Key to Species of Oreohelix, he altered these definitions as follows: “Embryonic shell strongly ribbed radially, of about Wi whorls . . . Subgenus Radiocentrum; Embryonic shell of about 2% whorls, which are striate and usually with spiral lines . . . Subgenus Oreohelix s. str."" Although the definition of Radiocentrum, i.e., embryonic shell of one and one-half whorls and strongly ribbed, can be applied to fossil Oreohelicidae without modification, the criteria for Oreohelix essentially denies fossil existence of the genus. Nearly all fossil Oreohelicidae have been reclassified as Radiocentrum (Roth, 1986; Tay- lor, 1975), I know of no fossil oreohelicid with a protoconch of two or more whorls. Considering modem distributional patterns, it is unlikely that the genus Oreohelix is a modem derivative of the genus Radiocentrum, although ancient derivation is plausible. Therefore, to be of any value in defining fossil taxa, the Pilsbry definition of Oreohelix (above as subgenus) must be altered to admit those taxa with a weakly ornamented (striate? or weakly costulate) protoconch of one and three-quarters or more whorls, and such definition is used herein. Oreohelix dawsonae Pierce, new species (Fig. 12.A-12.D) Diagnosis. — A strongly depressed heliciform shell, openly umbilicate, about 5.5 costate whorls, sharply angular at shoulder of juvenile whorls, becoming rounded in last whorl, aperture simple, strengthened within, basal lip weakly re- flected. Description. — Shell moderately large, strongly depressed heliciform to sub-lenticular, of 5.5 or more modestly rounded whorls, suture shallowly impressed, openly umbilicate. Embryonic whorls about 1.75, initial % whorl smooth to finely granulose, second Vs whorl with weak but regular wrinkle costulae, final Vi whorl with regular, but weak, costulae, becoming less regular, prosocline costae on neanic whorls; whorls initially strongly angular at shoulder, becoming rounded in last whorl, descend- ing in last few millimeters behind aperture. Aperture rounded, lip simple, blunt, oblique, strengthened within, columellar margin reflected around umbilicus, basal margin very weakly reflected. Measure- ments of holotype: height (est.) 8.0 mm, crushed; width 15.9 mm; HAV (est.) 0.50; whorls >5.5. Discussion. — The clear-cut distinction between the genera Oreohelix and Ra- diocentrum as proposed by Pilsbry (1939:418) appears blurred when dealing with mid-Tertiary species. Fossil Oreohelix are rare. Late Oligocene Oreohelix brandi Pierce, 1992, from the Cabbage Patch fauna of southwest Montana, has two or fewer, usually about 1.75, embryonic whorls, is initially smooth for first one-half whorl, then weakly costulate (Pierce, 1992:616). Roth (1986:249-252) has de- scribed two new species of the Oreohelicidae, as genus Radiocentrum, from Eo- cene to early Oligocene strata in the Three Forks Basin of west central Montana. Radiocentrum taylori Roth, 1986, has a “Protoconch consisting of 1.5 whorls, nuclear tip smooth, thereafter with prominent, elevated radial ribs . . . . ” Detail of the apical (protoconch) sculpture (Roth, 1986: fig. 20) clearly verifies that this is, according to the criteria established by Pilsbry (loc. cit.) a Radiocentrum. On the other hand, ‘'Radiocentrum’' laevidomus Roth, 1986, has a “Protoconch of 1.75 whorls, nuclear tip smooth, followed by smooth, regular, forwardly convex radial ribs . . . . “ Examination of Roth’s (1986:fig. 30) excellent photograph of Paratype USNM 377388, a hatchling young of R. laevidomus, shows that the smooth nuclear tip is about 0.75 whorl, followed by weak but regularly spaced costulae (wrinkles?) for the remainder of the embryonic whorls. The weak sculp- ture and extended protoconch (1.75 whorls) meet the criteria established above for the genus Oreohelix, and compares very closely with the character of embry- 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 51 onic whorls of both Oreohelix brandi and Oreohelix dawsonae. From this, it would appear that ''Radiocentrum” laevidomus is best considered an Oreohelix. Oreohelix dawsonae is quite similar to O. laevidomus^ but differs in being larger, and in having an even more depressed spire (H/W est. 0.50 versus HAV 0.55 to 0.63 for O. laevidomus). Oreohelix brandi, although about the same size, has a much more elevated spire (HAV —0.77). Oreohelix leconti (Steams, 1902) (holotype not seen), from the John Day Miocene of Oregon, is the only other fossil member of this genus. It differs in being smaller, having a somewhat more elevated spire, HA¥ —0.55, and fewer whorls that are “closely and conspicuously ribbed” (Stearns, 1902:154). Environment and Distribution. — An intermontane genus, widely distributed from Alberta and British Columbia to New Mexico, Arizona and Nevada, but generally north of the area occupied by Radiocentrum. They are generally found on limestone slopes and talus. In the northern part of their range they are found in welFwooded environments near streams and at low altitudes, and usually oc- cupy sparsely vegetated higher altitudes near their southern limits (Pilsbry, 1939: 412-540; Baquaert and Miller, 1973:32-37.) The earliest fossil Oreohelix are late Eocene-Oligocene O. laevidomus from the Three Forks Basin of west central Montana and Oligocene to early Miocene O. brandi from the Cabbage Patch fauna of southwestern Montana. Etymology. — Named to honor Dr. Mary Dawson, paleontologist, friend, and Curator of Vertebrate Paleontology, Carnegie Museum of Natural History, with whom I have spent many pleasant hours collecting fossils in the Kishenehn Basin. Material. — Seven specimens. Holotype CM-41624 and paratypes KUMIP-289,708 and KUMIP- 289,709 are from E-5 (NFF-1) locality, British Columbia, Kishenehn Formation, middle Eocene-late Oligocene. Additional paratypes are: CM-41625 from Commerce Creek (CMC-1) locality, British Columbia and PORH-111 (3 fragments) from Kintla Creek River (NFF-1 7) locality, Flathead County, Montana, all from Lower Member, Kishenehn Formation, late Eocene-Oligocene. Genus Radiocentrum Pilsbry, 1905 Type species. — Oreohelix chiricahuana Pilsbry, 1905. Radiocentrum kintlana Pierce, new species (Fig. 12.E-12.G) Diagnosis. — A moderate- sized, strongly depressed trochoidal shell with regu- larly spaced costae crossed by weak, discontinuous, spiral striae above periphery on ultimate whorl, strongly shoulder carinate in early whorls, more roundly keeled in last half whorl. Description. — Shell strongly depressed trochoid, of about 4.5 to 5.0 slightly rounded, sharply shoul- der carinate whorls that become more roundly keeled on last half of ultimate whorl, suture lightly impressed, openly umbilicate. Embryonic whorls about 1.5, with smooth nuclear tip preceding fine but strong, evenly spaced, prosocline costulae, subsequent whorls bearing coarser, regularly and closely spaced costae, with six discontinuous weak spiral striae crossing and cutting the costae into long granules above the periphery on the ultimate whorl. Aperture sub-lunate, simple, lip thin but blunt, not thickened within, not reflected. Measurements of holotype: height (est.) —7.0 mm (crushed); width 14.5 mm; H/W (est.) 0.48; whorls >4.5. Discussion. — ^The holotype, CM 41626, is fragmented, lacks the embryonic whorls and the palatal portion of the aperture. The paratype, CM 41627, is slightly crushed and incomplete, retaining only —4 whorls, but with good embryonic whorls. There is no hesitation in placing these specimens in the genus Radiocen- trum. Among modern members of this genus. Radiocentrum avalonense “Hem- phill” Pilsbry, 1905 is quite close, differing in having stronger embryonic cos- 52 Annals of Carnegie Museum VOL. 70 Fig. 12.— STYLOMMATOPHORA: OREOHELICIDAE. A-D: Oreohelix dawsonae n. sp. A-C: Ho- lotype (CM 41624); A. apical, X3; B. umbilical, X3; C. nuclear whorls, enlarged, X12; D. Paratype (KUMIP 289709), X3, compressed aperture; E-E Radiocentrum kintlaense n. sp. Holotype (CM 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 53 tulae, stronger and more closely spaced rough, irregular, costae, and in having spiral striae on the base only. Among fossil members of this genus that are similar, late Cretaceous Radiocentrum anguliferum (Whiteaves, 1885) is almost flat dor- sally, and Eocene R. grangeri (Cockerell and Henderson, 1912) is much larger, and both have a more angular periphery on the body whorl. Paleocene Radi- ocentrum thurstoni (Russell, 1926), early Oligocene Radiocentrum taylori Roth, 1986, and Oligocene Radiocentrum hendersoni (Russell, 1938) all have higher, more dome-like spires than R. kintlana. Environment and Distribution. — A calciphilic genus, usually found in isolated mountain top colonies, on talus slopes, rockfalls, and/or near water usually above 5,000 feet. Isolation and dispersion of species suggests relic populations repre- senting a previously more widespread distribution. Several species are known only from dead shells (Pilsbry, 1939:540-553). The genus is found living only in southern New Mexico and Arizona, northwestern Chihuahua, Baja California Sur, and, with a single species, Santa Catalina Island, California (Roth, 1986:248; Hochberg et ah, 1987). Fossil species are not uncommon, ranging from Cretaceous R. angulifera (Whiteaves, 1885) and Paleocene R. thurstoni (Russell, 1926) of western Alberta, through several mid-Tertiary taxa of the intermontane West. Ra- diocentrum taylori Roth, 1986, from the late Eocene to early Oligocene of the nearby Three Forks Basin of west central Montana, is temporally and geograph- ically close. Etymology. — Named in reference to the Lower Kintla Rapids locality where this species was found. Material. — Two specimens. Holotype, CM-41626, and paratype, CM-41627, are from Lower Kintla Rapids (NFF-18) locality, Flathead County, Montana, Lower Member, Kishenehn Formation, late Eocene-Oligocene. Family Megomphycidae Baker, 1930/? Genus Polygyroidea Pilsbry, 1923 Type species. — Daedalochila harfordiana Cooper, 1870. Polygyroidea montivaga Pierce, 1992 (Fig. 12.H-12.J) Discussion. — This species, so commonly encountered in the late Oligocene to early Miocene Cabbage Patch molluscan local fauna of the intermontane basins near Drummond, Montana, was rather rarely encountered at only two localities. One specimen, KUMIP 289741, from the CMC-1 locality, was complete enough to retain the distinctive parietal lamella (Fig. 1 1 . J), so there can be no doubt as to identification. The remaining specimens were separated from the more common Polygyrella, cf. P. polygyrella (Bland and Cooper, 1861) on the basis of shape, P. montivaga is thicker (H/W greater), and on the character of the basal portion of the later whorls, e.g., P. montivaga whorls are smoother on the base, whereas those of jP., cf. P. polygyrella are lightly wrinkled. Pilsbry (1939:554-555) notes 41626), X3; E. apical; F. umbilical; G. R. kintlai (CM 41627) X3, apical. MEGOMPHICIDAE. H-I. Polygyroidea montivaga Pierce, 1992 Hypotype (CM 41672), X3; H. apical; I. umbilical; J. P. mon- tivaga Hypotype (KUMIP 289741) X3, apertural, showing parietal lamella; K-M: Polygyrella poly- gyrella (Bland and Cooper, 1863) K-L; Hypotype (CM 41674), X3; K. apical; L. umbilical; M. P. polygyrella Hypotype (PAMM-109), X2, radial teeth folded back showing casts of teeth in body whorl in-filling. 54 Annals of Carnegie Museum VOL. 70 that “There are indications that the Ammonitellinae are a very old group, for a long time declining, some genera now perhaps approaching extinction. All of the genera except Glyptostoma are monotypic, and the ranges of the species are nar- rowly limited, . . . characteristics of decadent groups.” Measurements of hypo- type: width 8.4 mm; height 4.7 mm; HAV 0.56; whorls >7.25. Environment and Distribution. — This currently monotypic genus, P. harfordi- ana, is found living only in Fresno and Mariposa Counties of California in Tran- sition Zone forests generally above 1,800 m altitude (one colony at 1,220 m), where the Mean Annual Temperature (MAT) is about 10°C and the Mean Annual Precipitation (MAP) is probably >100 cm (Pierce, 1993:989). The only other fossil occurence of P. montivaga is the late Oligocene to early Miocene Cabbage Patch fauna of southwestern Montana (Pierce, 1992). Roth (in press) describes a new species from the Cretaceous (Campanian) of California. Photographs fur- nished by Roth (1999) leave no doubt as to its generic identity. Pilsbry’s com- ments, above, are certainly substantiated. Material. — Nine specimens. Figured specimen CM 41672, KUMIP-289,742 (2), and PAMM-108 are from Wurtz Bend (NFF-19) locality, Flathead County, Montana. Figured specimen KUMIP 289,741, CM-41673 (2), and PAMM-107 (2) are from Commerce Creek (CMC-1) locality, British Columbia. Both localities are in the Lower Member, Kishenehn Formation, late Eocene-Oligocene. Genus Polygyrella Bland and Cooper, 1861 Type species. — Helix polygyrella Bland and Cooper, 1861. Polygyrella sp., cf. P. polygyrella (Bland and Cooper, 1861) (Fig. 12.K-12.M) Polygyrella sp., cf. P. polygyrella, B. Roth, 1986:253, figs. 31,32. Polygyrella polygyrella, G. D. Hanna, 1920«:3. Polygyrella indet., D. W. Taylor in C. P. Ross, 1959:71. Zonitoidesl sp., L. S. Russell, 1956«: 1 10,1 1 1, pl.l, fig 8. Description. — “Shell subdiscoidal with broad umbilicus contained approximately three times in diameter. Spire flat to low-convex, suture lightly impressed. Whorls tightly coiled, sometimes with closely spaced, forwardly convex radial grooves extending outward from suture, not reaching shoulder of whorl. Body whorl with up to four shallow transverse constrictions. Last whorl not markedly descending except immediately behind aperture. Aperture oblique, peristome thickened and slightly expanded outward” (Roth, 1986:253). Within the last whorl there are one or more radial rows of three teeth each. Discussion. — Referring to this taxon as P. sp., cf. P. polygyrella merely sig- nifies that, conchologically, the material at hand cannot be distinguished from the normal range of variation of shells of modem P. polygyrella. There is no con- notation that they are the same species, which is considered highly unlikely. Even minimal genetic drift, considering the years involved, would insure specific sep- aration. The shell material of these specimens is, essentially, original. Although the radial teeth are not visible through the shell, as with modem specimens, many of the fragmented specimens either have these radial teeth on detached shell frag- ments, or as molds on exfoliated cores. Figure 12.M (PAMM-109, Lower Kintla Rapids, NFF-18) is of a section of a whorl displaying as molds the arrangement of the three radial teeth, with each tooth laid back alongside its mold for com- parison. One specimen has a crushed but essentially complete aperture (PAMM- 108, NFF-1), which confirms Roth’s description, above. Neither Roth (1986:253), nor the author, were able to find any evidence of the large parietal tooth so characteristic of the modem species. Considering the number of specimens stud- 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 55 ied, by both Roth and the author, it appears most likely that this mid-Tertiary polygyrellid is without a parietal tooth, and is, therefore, specifically distinct from the modern species. That it also is found in the Three Forks Lf. and the White River Lf. suggests adaptation to a climate drier and warmer than the modem species lends some support to this theory. Better specimens are needed to confirm this. Russell (1956:110, 111, pi. 1, fig. 8), working with very limited and incomplete material from a single locality, E-2, considered apical fragments of P. sp., cf., P. polygyrella to be Zonitoidesl sp. A very limited collection was made from or near his E-2 locality, without finding any specimens referable to this species. Environment and Distribution. — Polygyrella is now a monotypic genus whose current range is limited to northern Idaho, Montana, and northeastern Washington, where it is found in damp spmce forest and on lava and schist talus (Pilsbry, 1939:559). As a fossil, the genus has been reported from strata as old as Creta- ceous, [P. parvula (Whiteaves, 1885), and P. amplexus (Meek and Hayden, 1857)] and as P. sp., cf. P. polygyrella from the Eocene by Roth (1986:253), and from the Miocene by Hanna (1920^?:3). I can find no reference to P. polygyrella as a Recent or Pleistocene fossil. Material. — More than sixty specimens. Figured Specimens are as follows: CM 41674 from Star- vation Bend (NFF-12) locality, and PAMM 109 from the Lower Kintla Rapids (NFF18) locality, both Flathead County, Montana. Additional specimens are as follows: PAMM 110 and 111 from Commerce Creek (CMC~1) locality (16); PAMM 1 12 and PAMM 1 17 from E-5 (NFF-1) locality (14), both British Columbia; KUMIP 289,744 (4) and PAMM 113 (9) from Starvation Bend (NFF-12) locality; PAMM 115 from Kintla Creek-River (NFF-17) locality (3); KUMIP 289,743 (1) and PAMM 114 (3) from Lower Kintla Rapids (NFF-18) locality; CM 41675 (8) and PAMM 116 (3) from Wurtz Bend (NFF- 19) locality, all Flathead County, Montana. All localities are in the Lower Member, Kishenehn For- mation, lower Eocene-Oligocene. Superfamily Helicoidea Rafinesque, 1815 Family Humboldtianidae Pilsbry, 1939 Genus Skinnerelix Evanoff and Roth, 1992 Type species. — -Helix leidyi Hall and Meek, 1855. Diagnosis. — “Shell large (adult shells larger than 2 cm in maximum diameter); globose-conic; height to maximum diameter ratio greater than 0.8; embryonic shell smooth, consisting of first 1.2 whorls; neanic sculpture of growth rugae and coarse, somewhat crude granulation arranged in diagonal rows; last whorl broadly rounded, tumid, descending, constricted basally just behind lip; peristome flaring, narrowly turned outward, reflected at base and columella; base narrowly, obliquely perforate” (Evanoff and Roth 1992:124). Environment and Distribution. — Humboldtiana von Ihering, 1892 “is a plau- sible environmental analog to Skinnerelix . . . ranges from Mexico City, Mexico, north to the Guadeloupe Mountains in southeast New Mexico ... It typically lives in a variety of substrates and woody vegetation, ranging from oak forest on lime- stone to high coniferous forests and mixed scattered woodlands on volcanic rocks (Pilsbry, 1939). The climate in the range of Humboldtiana is subtropical . . . with mean annual range of temperatures of 4-20°C ...” (Evanoff and Roth, 1992: 129). Fossil Skinnerelix leidyi are known from the late Eocene and early Oligo- cene (Chadronian and Orellan Land Mammal Ages) of Nebraska, South Dakota, Utah and Wyoming (Evanoff and Roth, 1992:fig. 8). 56 Annals of Carnegie Museum VOL. 70 Skinnerelix rothi Pierce, new species (Fig. 13.A-13.C) Diagnosis. — Shell large, openly umbilicate, nuclear whorls smooth to finely pitted, sculpture of rough collabral costae broken into irregularly elongate, flat- topped granules, body whorl bears incised, spiral lines below and parallel to suture. Description. — Shell large, globose-conic, openly umbilicate, with about 4.75 rapidly expanding whorls; spire broadly convex, nuclear diameter 0.65 mm, embryonic whorls 1.5 to 1.75, smooth to finely pitted, flatly rounded, suture modestly impressed, indistinctly separated from subsequent whorls initially bearing closely spaced, irregular, collabral costellae which gradually become rough collabral costae generally broken into irregularly elongate, flat-topped granules aligned along costae; body whorl with indistinct, incised, spiral lines below and parallel to suture, last Vs whorl behind aperture descends rather abruptly; aperture lunate-ovate, outer lip thickened and narrowly reflected, basal and columellar becoming recurved. Measurements, estimated, of compressed holotype: height 19 mm; width 24 mm; HAV 0.79; height of aperture 13 mm; width of aperture 14 mm; whorls 4.75. Discussion. — Skinnerelix rothi is obviously very similar to S. leidyi, from which it is not easily separated. The chief differences in S. rothi are a slightly more depressed shape, an open umbilicus, incised, spiral lines parallel to the suture on the ultimate whorl, and costae broken into irregularly elongate, flat-topped gran- ules on the ultimate whorls, which are quite different from the fine, diagonally arranged, collabral granulations of S. leidyi. Skinnerelix rothi, as well as S. leidyi, can be quickly distinguished from various species of the genus Humboldtiana von Ihering, 1892 by the thickened, reflected, and recurved lip typical of Skinnerelix. Roth (1999, personal communication) kindly examined these, and other large gastropods from this fauna, and his keen observations and comments are reflected in the Description above. Etymology. — rothi, in honor of Barry Roth, University of California, Berkeley, Museum of Pale- ontology, and Editor of The Veliger, a preeminent molluscan paleontologist. Material. — More than thirty-six specimens, all incomplete, badly compressed and/or crushed and fragmented. The holotype, figured specimen CM 41676, is from Starvation Bend (NFF12) locality, Flathead County, Montana, Kishenehn Formation, Coal Creek member, late Oligocene. Additional specimens: Paratypes CM 41677 (6), KUMIP 289,745, KUMIP 289,746 (6), and PHBD 101-105(23) are from Starvation Bend (NFF-12) locality, PHBD- 106 is from Moose City South (NFF-10) locality, and PHBD 107 (3 apical fragments) is from Lower Kintla Rapids (NFF-18) locality, all Flathead County, Montana, Lower Member, Kishenehn Formation, late Eocene-Oligocene. Superfamily Xanthonychoidea Strebel and Pfeiffer, 1880 Family Helminthoglyptidae Pilsbry, 1939 Genus Xerarionta Pilsbry, 1913 Type species. — Arionta veatchii Tryon, 1866, ex Newcomb MS. Xerarionta constenii Pierce, new species (Fig. 13.D-13.F) Diagnosis. — Shell moderately large, depressed globose, obliquely umbilicate, aperture roundly lunate, sculpture of embryonic whorls smooth to finely granular, subsequent whorls with collabral costellae that become rugae extending from su- ture to umbilicus on ultimate whorls. Description. — Shell moderately large, depressed globose, spire convex, apical angle about 130°, umbilicus obliquely open, nuclear diameter 0.65 mm, embryonic whorls about 1.5, smooth to very finely granular, neanic whorls bear weak, closely spaced, collabral costellae becoming irregularly spaced rugae that continue over periphery into umbilicus on ultimate whorl, aperture roundly lunate, scarcely descending, outer and basal lips missing, columellar lip thin, broadly recurved around um- 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 57 Fig. 13.— STYLOMMATOPHORA: HUMBOLDTIANIDAE. A-C; Skinnerelix rothi n. sp. A-B: Ho- lotype (CM 41676), X3; A. apical; B. umbilical, note Anguispira sp., cf. A. alternata wedged in aperture; C. Skinnerelix rothi Paratype (PHBD-102), X1.7, detached aperture, frontal view. HEL- MINTHOGLYPTIDAE. D-F; Xerarionta constenii n. sp. Holotype (CM 41678), X3; D. apical; E. umbilical; F. apertural. 58 Annals of Carnegie Museum VOL. 70 bilicus. Measurements of holotype; width, major 18.2 mm; width, minor 15.1 mm; height 14.3 mm; HAV 0.79; width of aperture 10 mm; height of aperture 10 mm; whorls >4.2. Discussion. — Roth (personal communication) examined the holotype and con- cluded that this specimen is immature. The few whorls of the holotype are in agreement with this conclusion. Roth further compared the holotype to living X. redimita (W. G. Binney, 1858), from San Clemente Island, and found similarities in shape, but differences in sculpture. Fossil Helminthoglypta bozemanensis Roth, 1986, from the nearby late Eocene-early Oligocene Three Forks Basin is grossly similar but is smaller and has a more elevated spire. Fossil Xerarionta are not well known. Late Eocene (Uintan-Chadronian) Xerarionta waltmilleri Roth, 1984, is quite similar to X. constenii, especially in surface sculpture, but is much larger. Evanoff and Roth have another, yet undescribed, Xerarionta from latest Chad- ronian and/or early Orellan White River strata near Douglas, Wyoming. The two species can be discriminated by the difference in dimensions of the nuclear whorls (Roth, 1997, personal communication). Environment and Distribution. — Roth (1984:214) found that "'Xerarionata now occurs on the southern California Channel Islands and the adjacent mainland, and along the western part of Baja California between latitudes 24°-31°N. The Baja California occurrences are all within the arid, subtropical to tropical Sonoran Desert.” Roth (op. cit.) further determined the climatic conditions of known pop- ulation localities ranged from MAT of 17°-21.5°C with mean annual ranges of temperature from 7.2°-10°C, and generally are within areas influenced by the Pacific fog. Geologic range of the genus includes late Eocene to early Oligocene (Uintan-Orellan). Etymology. — Named to honor Kurt Constenius, geologist, friend and driving force behind this mul- tidisciplinary study of the Kishenehn Basin. Material. — Thirteen specimens. Holotype: CM 41678, is from Bowman Creek (BWN-2) locality, and paratypes: CM 41679 (2); KUMIP 289747; KUMIP 289748 (2); PHLM-101; PHLM-102 (3); and PHLM-105 are from Starvation Bend (NFF-12) locality, both Flathead County, Montana. Additional paratypes: PHLM-103 (4 apical fragments) is from E-2 (NFF-6) locality; and PHLM 104 (4 apical fragments) is from E-4 (NFF-2) locality, both British Columbia. All localities in the Lower Member, Kishenehn Formation, late Eocene-Oligocene. Aquatic Taxa Order Mesogastropoda Superfamily Rissooidea Gray, 1847 Family Hydrobiidae Troschel, 1857 Discussion. — The Hydrobiidae are among the many freshwater snail families wherein even generic identification on the basis of shell features alone is risky. As Taylor (1975:379) so aptly stated, “Classification of Hydrobiidae is based mainly on features of the operculum and reproductive system, for shells of these snails are generally without distinctive features.” Although it is believed that assignment to subfamilies herein is correct, generic assignments are less certain. Subfamily Hydrobiinae s. str. Genus Tryonia Stimpson, 1865 Description. — “Shell perforate, elongated, turreted, subulate, acute at summit amd rather pointed at base; surface longitudinally ribbed or plicated, not spinous; whorls numerous, shouldered. Aperture Table 3,—Kiskenehn molluscan fauna, aquatic, by locality. X indicates type locality, or locality of figured specimen. Entry in columns to the right of BWN- 2 if taxon is previously known; CPch = Cabbage Patch; 3Fks = Three Forks; RSL = RusselVs Kishenenh fauna. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 59 VO ^ feo z,x o [i, fez |S p Ei 2 fcU feS feu Z2 few |S d a ^ § I C<5 ^ p s la OkO ^ ■5« o s X X X X X X XX X X X X X X X XX XXX X XXX X XXX X X X XXX X X XX X XXX X X X X X X X U O k. s it a u I S d c s ■8 s 1^ t O Q s g « 1^- b w ^ ^ ^ e I 5^ I s s g 0 e 1 a e s S ^ 15 9 o Q g ^ I i ^ I K ^ X X XX XX X X XXX X X X BIVALVES Sphaerium progrediens Sphaerium (M.) discus 60 Annals of Carnegie Museum VOL. 70 small, oblique, rhombo-ovate; and somewhat pointed, sinuated and effuse at base; outer lip thin and sharp, projecting below; inner lip appressed to the whorl above, peritreme however continuous” (Stimpson, op. cit.). Type species. — Tryonia clathrata Stimpson, 1865. Tryonia russelli Pierce, new species (Fig. 14.F-14.H) Goniohasis sp. Russell, 1952:125, 132, pi. 19, fig. 14-17; 1956:105. Diagnosis. — Shell small, slender, with strongly rounded whorls, blunt apex, very weak spiral striae and collabral growth wrinkles becoming rugae on last whorls. Description. — Shell large for Tryonia, thin, minutely perforate, very slender, whorls slowly and regularly enlarging, about 20°, from an obtuse apex; nucleus flat or scarcely projecting, first two whorls smooth, subsequent whorls with irregular collabral growth wrinkles becoming noticeably costate on ultimate whorls, all whorls well rounded, suture deeply impressed; aperture ovate-lunate, simple, thin, reflected over columellar margin. Measurement of holotype: height, estimated about 9.0 mm; width, 2.7 mm; HAV 3.33; height of aperture, 2.6 mm; width of aperture, 1.7 mm; whorls estimated 7.0. Discussion. — Had Russell (1952:132) seen a more complete specimen, he would have realized that this little snail is a hydrobiid, not a pleurocerid. The thinness of the shell, the height and inflation of the whorls, and the blunt apex all confirm this. Among the seven species of this genus previously known (Leon= ard and Franzen, 1946), the most strikingly similar are T. hibbardi Leonard and Franzen, 1944, from the early Pliocene Laverne Formation of northwestern Oklahoma, and T. ellipsostoma Pilsbry 1934a; 1934Z?, from the early Pliocene Tulare Formation of the Kettleman Hills, California. Tryonia russelli differs mark- edly from both in size. However, the three species appear to be closely related and to form a natural group, with T. russelli the largest species of this group, as well as the oldest. Etymology. — Named to honor the late Dr. Loris S. Russell, Royal Ontario Museum, who first en- countered this species during his pioneering work in this valley. Environment and Distribution. — Pilsbry (1934a:542) interpreted the type lo- cality of Tryonia ellipsostoma (Kettleman Hills, California) to have been a large, shallow lake with abundant aquatic flora. Leonard and Franzen (1944:34), after an extensive review of the total Laveme fauna (Beaver County, Oklahoma), reached a quite similar conclusion, either a large, shallow, well-vegetated lake, “or more probably, a series of small lakes or ponds in a large basinal area and fed by freshwater streams.” Both are of early Pliocene age. Material. — Four nearly complete specimens, variously crushed, and six detached apical segments of 2-3 whorls each. Holotype, CM 41680, and paratypes CM 41681, KUMIP 289,749 and KUMIP 289,750, are from W-2 (NFF-5) locality, British Columbia, Lower Member, Kishenehn Formation, late Eocene-Oligocene. Subfamily Nymphophilinae Taylor, 1966 Genus Cincinnatial Pilsbry, 1891 Type species. — Paludina integra Say, 1821. Cincinnatial bowmanana Pierce, new species (Fig. 14.D, 14.E) Diagnosis. — A small, straight- sided conical shell of about 5.5 whorls, with tiny nucleus elevated above subsequent whorl. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 61 Description. — A small, conical shell, apical angle about 55°, of more than 5.5 well rounded, reg- ularly expanding whorls; nucleus tiny, about 0.20 mm in diameter, elevated above second whorl, apex acutely rounded, embryonic whorls about 1.5, finely punctate, with very weak collabral wrinkles after second whorl, strongest on body whorl; unbilicus tiny, chink-like, aperture ovate, a little more than one-third of shell height, acutely angular at parietal insertion, adnate to body whorl, basal and outer lip simple and sharp, columellar margin slightly flared. Measurement of holotype; height, 4.1 mm; width, 2.55 mm; W/H 0.62; height aperture; 1.5 mm; width aperture, 1.3 mm; whorls 5.7. Discussion. — Provisional placement of this species in the genus Cincinnatia, which is generally of eastern distribution, rather than the closely allied and con- chologically almost identical genus Pyrgulopsis Call and Pilsbry, 1886, which is mostly western in distribution, is based on the projecting nucleus and acutely rounded apex of this species. Cincinnatia cincinnatiensis (Anthony, 1840), which most closely resembles this new species, is distributed as far west as Utah (Clarke, 1973:242). Cincinnatia cincinnatiensis can be distinguished from C?. bowmanana by its less acute (70°), convexly conical spire and slightly taller aperture. Although several fossil species resemble Cl. bowmanana, they can be differentiated by the unique combination of characters possessed by Cl. bowmanana, i.e., projecting nucleus, acutely rounded apex, straight- sided conical shape, apertural character and tiny, chink-like umbilicus. Among those with generally similar appearance are Cretaceous Hydrobia subconica Meek, 1876, Eocene H. anthonyi (Meek and Hayden, 1856), and Melania simpsoni (Meek, 1860), Pliocene Amnicola bithno- ides Yen, 1944, and Pleistocene to Recent Pyrgulopsis longinqua (Gould, 1855). Cincinnatial bowmanana does not appear to be the ''Goniobasis'' sp. collected by Russell (1952:132) from his locality W-1. Russell’s material was fragmentary and of too poor quality to describe in detail, however it seems to be a more elongate species, with an apical angle of —35° (Russell, 1952:132, pi. 19, figs. 14-16). Etymology. — Named for the type locality, Bowman Creek, a tributary of the North Fork of the Flathead River in Flathead County, Montana. Environment and Distribution. — Cincinnatia cincinnatiensis, which may be a close, living analog, is a denizen of diverse habitats ranging from large lakes to small rivers, in general shallow, but well vegetated, water and over bottoms rang- ing from mud to sand. The species is widely distributed, from New York to southern Saskatchewan and south to Utah and Texas (Clarke, 1973:242). The geologic range of Cincinnatia is Pleistocene to Recent, and for the companion genus, Fonticella, Gregg and Taylor, 1965, early to middle Pliocene to Recent. Material. — Two specimens. Holotype, CM 41682, a complete shell with some apertural breakage, and paratype CM 41683, an apical fragment of three whorls, and a smaller fragment, are from Bowman Creek (BWN-2) locality, Flathead County, Montana, Lower Member, Kishenehn Formation, early Eocene-Oligocene. The matrix mold of the back of the holotype is retained as PHDB-101. Subfamily Lithoglyphinae Hartmann, 1821 Genus Fluminicola Stimpson, 1865 Type species. — Paludina virens Lea, 1838. Fluminicolal calderense Pierce, new species (Fig. 14.A-14.C) Diagnosis. — A small, bluntly conical shell with small, non-protruding nucleus, large, subrectangular aperture with columellar margin expanded over and around umbilicus. Annals of Carnegie Museum VOL. 70 Fig. 14. — MESOGASTROPODA: HYDROBIIDAE. A-C: Fluminicolal calderensis n. sp. XIO; A- B: Holotype (CM41684), A. apertural; B. lateral; C. Paratype CM 41685, apical; D-E: Cincumatial 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 63 Description. — Shell small, bluntly conical, with four or more rounded whorls, initially regularly expanding, body whorl expanding rapidly, suture well impressed; nucleus small (0.25 mm) on plane of second whorl or very slightly elevated, embryonic whorls about 1.5, smooth, subsequent whorls shiny with very faint collabral wrinkles, strongest on body whorl, umbilicus narrow, covered by peristome, aperture subrectangular, parietal insertion of lip at near right angle to body whorl, peristome free of body whorl, thickened by inner callus, columellar lip thickened, reflected over and around umbilicus, basal and outer lip simple, thickened within, edge slightly blunt. Measurements of holotype; height, E3.4 mm, spire missing; width, 3.3 mm; height aperture, 2.0; width aperture, 1.8 mm; whorls unknown. Measurements of paratype: height, E3.0 mm, crushed; width, 2.9 mm; height aperture, 1.65 mm; width aperture, 1.0 mm; whorls —4.2. Discussion. — There is uncertainty with regard to generic assignment of this species. Fluminicolal calderense has many characteristics in common with both the genus to which it is tentatively assigned and to Somato gyrus Gill, 1863. Fluminicola was selected based on apertural characteristics, and the exclusively western distribution of this genus. Somatogyrus is, essentially, an eastern taxon. Among modem species, F?. calderense resembles both F. fusca (Haldeman, 1847), and S. amnicoloides Walker, 1915, but can be differentiated from F. fusca by its taller, convexly conic spire, oval aperture, and only slightly reflected col- umellar margin, and from S. amnicoloides by its much taller, convexly conic spire. Among fossil species, FI. calderense resembles some specimens of Pliocene F. kettlemanensis Pilsbry, 193 Aa,b from which it can be distinguished by the lower spire and more subrectangular aperture of F. kettlemanensis. According to Pilsbry (1934fi(:549), “There is no conchological difference between Lithoglyptus and Fluminicola, the distinction being in the form of the verge.” Etymology. — calderense, with reference to Calder Creek, a west side tributary of the North Fork of the Flathead River, near the type locality. Environment and Distribution. — Living species of genus Fluminicola are re- stricted to the United States west of the Continental Divide, ranging from the Canadian border south to northern California, Nevada and Wyoming. It is found in a variety of habitats ranging from spring pools to medium-sized rivers (Burch, 1989:102). The geologic range is unknown. Material. — Three specimens. Holotype CM 41684, spire missing, and paratype CM 41685, spire crushed, aperture damaged, are from W-2 (NFF-5) locality, British Columbia; additional specimen, PHDB-102, spire fragment, from Wurtz Bend (NDD-19) locality, Flathead County, Montana, all from Lower Member, Kishenehn Formation, late Eocene-Oligocene. Superfamily Valvatoidea Gray, 1840 Family Valvatidae Gray, 1840 Genus Valvata Muller, 1774 Type species. — Valvata cristata Muller, 1774. Valvata procera (Russell, 1952), new combination (Fig. 14.I-14.K) Gyraulus procerus Russell, 1952:131,132, fig. 10, pi. 19, figs. 3-13; Ross, 1959:70. Valvata procerus (Russell, 1952), Constenius et. al., 1989:197. bowmanana n. sp. Holotype (CM 41682), XIO; D. apertural; E. apical; F-H: Tryonia russelli n. sp. X5, F. Holotype (KUMIP 289748), X5, apertural; G-H; Paratype CM 41680 with a spire fragment from lot Paratype CM 41681; G. apertural; H. combined for reconstructed lateral view. VALVATIDAE. I-K: Valvata procera (Russell, 1952) Hypotype (CM 41686), XIO; I. apical; J. umbilical; K. apertural; L-N: Valvata procera spatiosa n. ssp. Holotype (CM 41688), XIO; L. apical; M. umbilical; N. ap- ertural. BASAMMATOPHORA: PLANORBIDAE. O-Q: Helisoma triangulata n. sp. Holotype (CM 41690), XIO; O. left; P. apertural; Q. right. 64 Annals of Carnegie Museum VOL. 70 Description-emended. — Shell small, depressed, of about three regularly expanding whorls, initial 1.2 whorls smooth, gradually acquiring costellae until, after about 1.75 whorls, sculpted with crowded fine, irregular and weak collabral costellae, most prominent on periphery; apex near planispiral for first 1.5 to 2.0 whorls, then subsequent whorls descend at increasing rate; whorls near circular, be- coming slightly and obliquely oval at aperture, peristome complete, narrowly appressed to penultimate whorl, umbilicus open, about 1/5 diameter. Measurements of hypotype are as follows: diameter, major, 3.0 mm; minor, 2.3 mm; height 1.7 mm, H/D 0.57; whorls, 3.1. Discussion. — This little valvatid is widely dispersed in the Kishenehn Basin, found at nearly all localities with suitable environments. Among living or more recently fossil valvatids, V. procera resembles V. humeralis Say, 1829, differing in being smaller, having fewer whorls, a more depressed shape, and more notice- ably oval aperture. Valvata humeralis is one of the western group of valvatids. Among fossil valvatids, V. procera appears to be closest to Oligocene V. paula Pierce, 1993, from which it differs in being more depressed, less sharply sculp- tured, having a less rapidly descending ultimate whorl, and an oblique, bluntly oval aperture. Russell (1952:132; 1955:104-105) reported this species from his W-1 (type locality), and W-3 (later changed to C-3) localities, and from an outcrop near the mouth of Whale Creek, west bank of the Flathead River in Flathead County, Montana. These localities have been destroyed by erosion. The figured hypotype (CM 41686), although slightly crushed and from a different locality, appears representative of his type specimens. Environment and Distribution. — Valvata humeralis^ which may be the closest modern analog, is a western species, found from British Columbia and Montana south to California and New Mexico and into central Mexico (Bequaert and Mill- er, 1973:213), and is fossil, from the same general area, from Pliocene to Recent (Taylor, 1966). Valvata, in general, are inhabitants of well- vegetated lakes or slow flowing rivers, with little apparent preference in bottom conditions (Clarke, 1973). The genus Valvata is ancient, known from Jurassic to Recent, with several species from the Eocene-Oligocene. Material. — More than 100 specimens. Figured specimen: Hypotype CM 41686 is from Bowman Creek (BWN-2) locality, Flathead County, Montana. Additional specimens, Hypotypes CM-41687 (18), KUMIP 289,751 (2), KUMIP 289,752 (18), PVLV-105 (4) from Bowman Creek locality; PVLV- 106 from Wurtz Bend (NFF-19) locality; PVLV-107 from Kintla Creek Carnegie (KTL-1) locality; PVLV-108 (10) from Starvation Bend (NFF-12) locality, all Flathead County, Montana; and PVLV- 109 (10) from Moose City North (NFF-9) locality; PVLV-110 (20) from Island (NFF-4) locality; PVLV-1 1 1 from E-4 (NFF-2) locality; PVLV-1 12 (20) from E-5 (NEF-1) locality; and PVLV-113 (20) from Commerce Creek (CMC-1) locality, all British Columbia. All specimens from Lower Member, Kishenehn Eormation, late Eocene-Oligocene. Valvata procera spatiosa Pierce, new subspecies (Fig. 14.L-14.N) Diagnosis. — A Valvata procera in which the rate of expansion of the whorls increases more rapidly, with a marked increase after —2.5 whorls. Description. — Shell moderately sized, depressed, of about 3.5 whorls, first 2.5 whorls like the nominative form, at which point the rate of expansion of the subsequent whorl increases markedly, aperture round to slightly oval, peristome entire, scarcely attached to penultimate whorl, or barely detached. Measurements of holotype: diameter, major, 5.0 mm; minor, 3.6 mm; height, 3.0 mm; H/D 0.60; whorls, 3.5. Discussion. — This valvatid is so unique, in size and shape (rate of expansion of whorls), when compared to the nominative species found at all other localities in the Kishenehn Basin that the first reaction is to consider it a different species. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 65 However, detailed comparison under higher magnification showed that the initial ±2.5 whorls are, although proportionately slightly larger, essentially identical. The marked distinction is in the even greater rate of expansion in the ultimate whorl. Considering the great polymorphism of some living valvatids, it is felt better to treat this taxon as a subspecies; however, whether it is, in fact, a true subspecies, or an ecophenotype, is impossible to ascertain. Although this taxon is found in close geographic proximity to the nominative form, limited faunal evidence sug- gests that the type locality may be of a slightly different age than adjacent local- ities. Valvata procera spatiosa bears a marked resemblance to fossil V. humeralis densistriata Pilsbry, 1934a;Z? from the Pliocene Kettleman Hills of California, differing in being somewhat larger, more depressed (HAV=0.60 versus H/ W=0.90), and in lacking the sculpture of fine threads. Etymology. — Latin, spatiosa, spacious, broad, large. Environment and Distribution. — See above under Valvata procera. Material. — Fourteen specimens. Holotype, CM 41688, and paratypes CM 41689, KUMIP 289,753, and PVLV 114, from W-2 (NFF-5) locality, British Columbia. Type horizon: Lower Member, Kishe- nehn Formation, late Eocene-Oligocene. Order Basommatophora Superfamily Ancyloidea Rafinesque, 1815 Family Planorbidae H. and A. Adams, 1855 Note. — Taxonomy of the Planorbidae herein generally follows that of Hubendick, 1978. Subfamily Planorbinae Pilsbry, 1934 Tribe Helisomatini Clarke, 1978 Genus Helisoma Swainson, 1840 Type species. — Planorbis bicarinatus Say, 1817. Subgenus Helisoma (5. str.) Helisoma triangulata Pierce, new species (Fig. 14.0-14.Q) Diagnosis. — A small, biconcave planorbid shell with three carina, one periph- eral, one ringing a steep, deep umbilicus, the other ringing a deeply depressed spire. Description. — Shell small, biconcave planorbid, of about four regularly increasing whorls, last 0.25 whorl increasing more rapidly; right side deeply and eccentrically depressed, sharply carinate on all whorls after embryonic whorls, sculpted with fine, closely spaced, prosocline costellae on last 1.5 whorls; left side deeply and steeply spirally umbilicate, sharply carinate at umbilical margin, sculpted with weak, straight radial costellae, third carina at periphery of whorls; aperture irregularly pentagonal, produced on right side, outer lip segments, between three keels, near straight on left side, broadly arched on right side. Measurements of holotype CM 41690: width 3.5 mm; height 2.3 mm; whorls est. 4. Range of paratypes: height 3. 8-5.0 mm; width 2. 3-2. 6 mm, mostly crushed; whorls about 4. Discussion. — The size and shape of this little species is so unique that it is not likely to be confused with any living species. It is a midget among the Helisoma and was initially considered to be a new genus of the Planorbidae, but further study of the specimens available showed that, despite its small size, it is really a Helisoma of the subgenus Helisoma {s. str.), and, in many ways, resembles some of the varieties of H. anceps Menke, 1830, which are known from strata as old as late middle Pliocene in Oregon (Taylor, 1966:35). It is, however, instantly differentiated by its small size, triple carina, deep spire depression and irregularly 66 Annals of Carnegie Museum VOL. 70 pentagonal aperture. Fossil Helisoma are not common, and are not well under- stood, due in part to confusing taxonomy, and, in part, to the lumping of most planorbid genera into the inappropriate genus Planorbis Muller, 1774. In size, H. triangulata compares with Pliocene H.{1) kettlemanense Pilsbry, 1934^;^?, but is instantly differentiated by the triple carina. Helisoma triangulata also bears a resemblence to modem Helisoma (Carinifex) newberryi (Lea, 1858), but can be easily differentiated by its size, medial keel and sunken apex. Etymology. — Latin, tri, combining form of three and angulata, feminine, angular, referring to the distinctive triple keels or carinae of this species. Environment and Distribution. — Helisoma anceps {s. lat.), which most closely resembles H. triangulata, is found in almost all permanent water situations with bottoms of all types and vegetation in various amounts. Distribution is from cen- tral and southern Canada, essentially throughout the United States, and into north- western Mexico (Clarke, 1973:430-432). Modem shells were collected on the Flathead River in the Kishenehn Basin. The geologic range of the genus Helisoma is difficult to establish, since the proper taxonomic position of the many early Cenozoic “Planorbis” is uncertain, but surely extends back to early Pliocene H. (?) kettlemanense. The modern analog of this species, H. anceps, is known from the late middle Pliocene to Recent on the High Plains (Pierce, 1975:126). Material. — Twenty-two specimens. Holotype, CM 41690 and all paratypes CM-41691 (6), KUMIP 289,754, KUMIP 289,755 (6), PPLN 113 (6), and PPLN-114 (2), are from W-2 (NFF-5) locality, British Columbia, Lower Member, Kishenehn Formation, late Eocene-Oligocene. Genus Planorbella Haldeman, 1842 Subgenus Piersoma Dali, 1905 Type species. — Planorbis trivolvis Say, 1816. Planorbella fordensis Pierce, new species (Fig. 15.D-15.G) Diagnosis. — A moderate- sized, shallowly biconcave, thick- walled planorbid, whorls rounded, without carina, sculpted by weak spiral threads and moderate costellae. Description. — Shell biconcave planorbid, of medium size, with about 4.25 expanding but modestly inflated thick-walled whorls of moderate axial height; right (umbilical) side shallowly and conically concave, hiding first whorl, whorls rounded, suture well impressed, sculpted with moderately strong prosocline costellae and weak spiral threads extending from suture to periphery, threads about six per mm on body whorl; left (spire) side very shallowly concave, whorls moderately rounded, suture impressed shallowly, all whorls visible, nuclear whorls about 1.5, smooth to finely punctate, subsequent whorls initially with fine, faint spiral threads, weak, shallowly prosocline costellae begin after second whorl, and are crossed by rare, discontinuous spiral threads, on body whorl, costae are irregularly interspersed among the costellae; aperture broken, but probably only slightly expanding. Measurements of holotype, an incomplete specimen: height >3 mm; width ElO mm; whorls >4. Discussion. — Unfortunately, only two incomplete specimens of this most in- teresting species were found; however, they are sufficient to recognize a new species. Planorbella fordensis most resembles modem P. subcrenulatum (Car- penter, 1856), a western species, in shell shape, sculpture, and, from fragmentary evidence, aperture, but it can be distinguished by a much shallower concavity on the left side, and the sculpture of both prosocline costellae and spiral threads. The oldest probable Piersoma fossil appears to be early Pliocene P. valens Leonard and Franzen, 1944, from the Laveme Formation of Oklahoma. Except for size. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 67 P. valens appears to be but weakly differentiated from P. trivolvis, found else- where in the latest Pliocene (Blancan) faunas east of the Continental Divide. Planorbella fordensis is grossly similar, but can be easily differentiated from P. trivolvis by its smaller size, and much finer sculpture and from both P. trivolvis and P. valens by the spiral threads. Etymology. — -fordensis, with reference to the type location near the mouth of Ford Creek, a tributary from the east of the North Fork of the Flathead River. Environment and Distribution. — Planorbella subcrenatum is extremely toler- ant, being found in swamps, ponds, streams, and rivers; mud substrates and abun- dant vegetation are preferred. Distribution western, from Yukon Territory to Cal- ifornia and east to Manitoba and Minnesota (Clarke, 1973:458). As with the genus Helisoma, it is difficult to establish the geologic range of the genus Planorbella, since the proper taxonomic position of the many early Cenozoic ''Planorbis” is uncertain, but it surely extends back to early Pliocene. Planorbella subcrenatum is known as fossil (as P. subcrenata) from two late Pliocene (Blancan) localities (Taylor, 1966:29, 70). Material. — Two specimens. Holotype: CM 41694. Type locality: South Ford Creek (NFF-22) lo- cality, Flathead County, Montana; Paratype CM 41695, an immature and/or broken back specimen of 2.25 whorls from E-3 (NFF-3) locality, British Columbia, both from Lower Member, Kishenehn For- mation, late Eocene-Oligocene. Genus Planorbula Haldeman, 1840 Type species. — Planorbis armigerus Say, 1821. Planorbula, cf. P. campestris (Dawson, 1875) (Fig. 15.K-15.M) Discussion. — This is one of two fresh-water taxa considered, although reluc- tantly, to be inseparable from a living taxon. However, as previously noted, there is no implication that they are specifically identical biologically, merely indistin- guishable conchologically, since there is no significant difference in shell mor- phology between these specimens and living specimens. The oldest P. campestris I have seen is from an early middle Pleistocene locality in North Texas (Pierce, 1975:41-44). Taylor (1966:83, 89) reported this taxon (also as cf.) from the mid- dle(?) Pliocene near Jackson Hole, Wyoming. The only differences observable between the Kishenehn specimens and excellently preserved early Kansan spec- imens from the North Texas locality (above) were a slight tendency to some angularity at the periphery, and only vague spiral striae. The figured specimen, CM 41696, although partly a steinkem, cannot be separated from modem speci- mens. Planorbula campestris can be easily separated from Biomphalaria hay deni Pierce, 1993, with which it coexisted locally, by the almost flat right side, the slowly and regularly enlarging whorls, and the conically concave left side. Plan- orbula campestris resembles Oligocene P. powelli Pierce, 1993, from the Cabbage Patch fauna of southwest Montana, but P. powelli has a slightly concave right side and an angular aperture that is deflected to the left. Measurements of figured hypotype, CM 41696: width 4.2 mm; height 1.6 mm; W/H 2.62; whorls 3.75. Range of specimens: width 3. 6-6. 2 mm; height 1.3-1. 7 mm; whorls 3. 6-4. 2. Environment and Distribution. — Clarke (1973:424) found P. campestris in gen- erally permanent waters such as swamps, flooded areas adjacent to ponds, roadside ditches, small ponds, and a slow-flowing small stream, all with mud bottom and abundant vegetation. It is a western species, found from British Columbia to VOL, 70 Annals of Carnegie Museum 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 69 Manitoba and south to Utah and New Mexico (Clarke, op. cit.). The geologic range of the genus Planorbula is uncertain, due to taxonomic uncertainty resulting from the use of “Planorbis'' for early Tertiary fossils, but it is known with cer- tainty from the late Oligocene (F. powelli) of southwestern Montana, and from the middle Pleistocene of Texas (Pierce, 1975:128), and as P. cf. P. campestris from the middle to late Pliocene (Blancan) of Wyoming (Taylor, 1966:89). Material. — Eleven specimens. Figured specimen, CM 41696, from Bowman Creek (BWN-2) lo- cality, Flathead County, Montana. Additional specimens CM 41697 (2), and KUMIP 289,757 (2), from Bowman Creek (BWN-2) locality, KUMIP 289,758 (2) from Wurtz Bend (NFF-19) locality, and PPLN-115 (3) from Starvation Bend (NFF-12) locality (2), all Flathead County, Montana. PPLN-116, one specimen in matrix is from Commerce Creek (CMC-1) locality, British Columbia. All locations are in the Lower Member, Kishenehn Formation, late Eocene-Oligocene. Genus Menetus H. and A. Adams, 1855 Type species. — Planorbis opercularis Gould, 1847. Subgenus Menetus s. str. Menetus textus Pierce, new species (Fig. 15.H-15.J) Diagnosis. — A small planorbid with a flat right (umbilical) side and a shallow, widely concave left (spire) side, sculpted with fine, close costellae and more widely spread, cross-cutting, spiral lines, aperture deeply lunate-roundly ovate. Description. — Shell small, planorbid, of about 4.5 rounded whorls, initial 1.5 to 1.75 whorls smooth; right side flat, whorls flatly rounded, suture moderately impressed, sculpture after embryonic whorls of fine, closely spaced, prosocline costellae cut by fine, more widely spaced, spiral lines, costellae becoming coarser and more widely spread in last 0.25 whorl; left side widely and shallowly concave, all whorls visible, whorls well rounded, embracing preceding whorl about 50%, sculpted with radial to orthocyrt costellae cut by fine, more widely spaced, spiral lines, costellae becoming coarser in last 0.25 whorl, interaction of costellae and lines give shell a fabric-like texture; aperture slightly expanded peripherally, deeply lunate-roundly ovate, advanced on right side, weak inner callus. Measurements of holotype, CM 41698: width, major 5.9 mm; width, minor 4.9 mm; height 1.6 mm; W/H 3.69; whorls 4.5. Range of paratypes: width 3. 8-7.0 mm; height 0.7-est. 1.7 (crushed); W/H 3.68-4.22; greatest whorls 4.6. Discussion. — Among modem species, M. textus appears related to the M. coop- eri (Baker, 1945) group (M. opercularis of Burch, 1989:202), especially those subspecies with rounded peripheries, from which it can be differentiated in having more, and more slowly expanding, whorls, in lacking the spiral ridges below the slight shoulder, and in lacking the tendency to malleation on the base. Considering the size, prominent spiral lines, and the textile-like texture of the shell of M. texta, the only fossil species with which it could be confused is lower Eocene ‘'Pia- norbis” storchi Russell, 1931 (most probably a Biomphalaria), which is twice as large, 13mm in width and 3.3 mm in height. Fig. 15.— BASAMMATOPHORA: PLANORBID AE. A-C: Menetus hilli n. sp. Holotype (CM 41692), X5; A. left; B. apertural; C. right; D-G: Planorbella fordensis n. sp. D-F: Holotype (CM 41694), X5; D. left; E. apertural; E right; G. P. fordensis paratype (CM 41695), XIO, apertural view of juvenile; H-J: Menetus textus n, sp. Holotype (CM 41698), X5; H. left; I. apertural; J. right; K-M: Planorbula sp, c.f., P. campestris (Dawson, 1875) Hypotype (CM 41696), XI 1; K. left; L. apertural; M. right; N-Q: Biomphalaria haydeni Pierce, 1993; N-P: Hypotype (CM 41700), XIO; N. left; O. apertural; P. right; Q. B. haydeni Hypotype (PPLN-108), XI 5, aperture showing basal and two palatal lamella, lower hollow. 70 Annals of Carnegie Museum VOL. 70 Environment and Distribution. — The suggested modem analog, Menetus coop- eri, is relatively common in California, Oregon, and Washington, and ranges north to Alaska and east to Alberta (Clarke, 1973:418). Clarke (op. cit.) collected spec- imens from a small stream and a moderate- sized lake, both with sand bottom and sparse vegetation. Previously known geologic range of the genus is from early Pliocene to Recent. Etymology. — Latin, textus, p.p. of texto, fabric, woven cloth, with reference to the surface texture of the shell. Material. — Eight specimens. Holotype, CM 41698 and paratypes CM 41699, KUMIP 289,759 (2), and PPLN-1 17 (1+3 fragments) are all from Wurtz Bend (NFF-19) locality, Flathead County, Montana, Lower Member, Kishenehn Formation, late Eocene-Oligocene. Subgenus Micromenetus F. C. Baker, 1945 Type species. — Planorbis dilatatus Gould, 1841. Menetus hilli Pierce, new species (Fig. 15.A-15.C) Diagnosis. — A small, lenticular, ultradextral species with spire in a spirally enlarging, near vertically walled pit, with near 90° shoulder as whorl expands to a roundly angular periphery, aperture roundly triangular, advanced on basal side. Description. — Shell small, lenticular, modestly hyperstrophic (ultradextral), of four or more rapidly expanding whorls; spire (left side) deeply depressed, in a spirally enlarging, near vertically walled pit, whorls strongly shouldered at about 90° departing pit, thence shallowly convex to a medial, roundly angular, periphery; embryonic whorls granulose, subsequent whorls with weak collabral wrinkles, becoming closely and weakly ribbed dorsally; base (right side) shallowly concave, whorls moderately overlapping, smoothly convex and faintly wrinkled between periphery and moderately impressed su- ture; aperture roundly triangular, moderately overlapping penultimate whorl on both sides, advanced on basal side. Measurements of holotype: height 1.8 mm; width, major 4.5 mm; minor 3.5 mm; whorls 3.75. Discussion. — Menetus hilli differs from Menetus textus in size, in that its left side (spire) is narrowly and relatively deeply sunken while its right side (umbi- licus) broadly and shallowly concave and in having a roundly angular, medial Carina on the ultimate whorl. Among living species of Micromenetus, it is closest to M. brogniatianus (Lea, 1842), which differs in having a near flat to only slightly concave right side, and an aperture produced markedly to the left side. Its roundly, but symmetrically, triangular aperture, appears unique among the members of this genus. Menetus hilli is also quite similar to Promenetus exacuous (Say, 1821), but lacks the flat spire, slightly descending aperture, more carinate periphery, and, commonly, a malleated appearance to the left side of P. exacuous. The exact relationship between the genera Menetus (s.l.) and Promenetus is un- clear and deserves study. Etymology. — Named to honor Warren Hill, fellow geologist, close friend and retired National Park Service Superintendent. Environment and Distribution. — Menetus {s. lat.) is not well known. Menetus {s. str.) is a western taxon, distributed from Alaska to California, but not often east of the Rocky Mountains. Micromenetus is eastern in distribution, from Maine to Florida, generally east of the Mississippi River, except for Iowa to Texas (F. C. Baker, 1945:182-190; J. B. Burch, 1989:202). The ecology of Menetus is not well known. Clarke (1973:418) suggests it to be a denizen of shallow streams and ponds, with sand bottoms, and sparse vegetation. Geologic range is unclear. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 71 probably due to confusion with Promenetus exacuous. E C. Baker (1945:187; 190) states “Pliocene to Recent.” Material. — Four specimens. Holotype, CM 41692, and paratypes, CM-41693 and KUMIP 289756 (2) from W-2 (NFF-5) locality, British Columbia, in Lower Member, Kishenehn Formation, late Eo- cene-Oligocene. Tribe Biomphalarini Hubendick, 1955 Genus Biomphalaria Preston, 1910 Type species. — Biomphalaria smithi Preston, 1910. Subgenus Tropicorbis Brown and Pilsbry, 1914 Type species. — Planorbis liebmani Dunker, 1850 {=P. havanensis Pfeiffer, 1839). Environment and Distribution. — Subtropical and tropical, generally in mud- bottomed freshwater marshes and bayous; they are well adapted to desiccation and capable of tolerating near freezing temperature briefly; distributed from South America, Caribbean, Central American and Gulf Coast of North America as far north as Baton Rouge, LA and Austin, TX (C. S. Richards, 1963). Tropicorbis is known from Oligocene [5. nebraskensis (Evans and Schumard, 1854) and B. havdeni Pierce, 1993] and, questionably in this subgenus, ?Eocene B. storchi, (Russell, 1931), (Pierce, 1993:984, 987). Biomphalaria hay deni Pierce, 1993 (Fig. 15.N-15.Q) Discussion. — These specimens are but slightly differentiated from those at the type locality in the late Oligocene (Arikareean) Cabbage Patch beds of south- western Montana (Pierce, 1993:983, 984), but the differences are so slight as to not suggest any separate taxonomic status. These slight differences include a wider range of angularity at the periphery, from almost round to quite sharp, a greater W/H in those with angular forms, a flatter right side, slightly more prom- inent costellae and an even greater frequency of irregular growth than observed within the type lots. These growth irregularities are usually due to abrupt changes in rate of coiling, leading to marked differences in overlap by succeeding whorls. Only one specimen displayed part of the apertural denticles, and, in it, the palatal (right side) lamellae are both slightly stronger than those observed in the type specimens. Extreme specimens with the most acute peripheries from the Kishe- nehn Basin, could, in gross appearance, be confused with some varieties of Pro- menetus exacuous (Say, 1821), but, even if the apertural lamellae are not apparent, could be easily separated from the flat to slightly domed whorls on the right side of P. exacuous. Measurements of the figured hypotype, CM 41700: width 4.6 mm; height est. (slightly crushed) 1.5 mm; W/H 3.07; whorls 4. Range of hy- potypes: width 3.4-5. 2 mm; height 1.1-est. 1.6; W/H 3.09-3.25; whorls 3. 5-4. 5. Material. — More than 60 specimens. Figured specimen, CM 41700 and CM 41701 (14), PPLN 119 and PPLN 120 (2) from Bowman Creek (BWN-2) locality, Flathead County, Montana. Additional specimens are as follows: PPLN- 120 (25) from South Ford Creek (NFF-22) locality; KUMIP 289,760 and KUMIP 289761 (14) from Wurtz Bend (NFF-19) locality; and PPLN-121 (8) from Starvation Bend (NFF-12) locality, all Flathead County, Montana; and PPLN- 122 from Moose City North (NFF- 9) locality, British Columbia. All localities are in Lower Member, Kishenehn Formation, late Eocene- Oligocene. 72 Annals of Carnegie Museum VOL. 70 Biomphalaria spira Pierce, new species (Fig. 16.A-16.C) Diagnosis. — A depressed biomphalarid with flat oval whorls and aperture, whorls scarcely overlapping, aperture not deflected from axis of shell. Description. — Shell a medium-sized, depressed planorbid, of about five or more slowly expanding oval whorls; right side slightly sunken, all whorls visible, whorls flatly rounded, scarcely embracing previous whorl, suture impressed, sculpture of weak, fine, regular, prosocline costellae; left side very shallowly and widely concave, all whorls visible, whorls flatly oval, suture moderately impressed, sculpture of weak, fine, regular, near radial costellae; aperture elongate and flatly oval, advanced on right side^ not deflected from axis of shell, last 0.5 mm slightly thickened and sculpted with more widely spaced costellae, bearing one or more teeth, a blunt, spirally elongated tooth is located about 0.5 mm behind aperture on left side. Measurements of holotype, CM 41702: width 9.0 mm; height crushed; whorls —5.3. Apertural measurements of paratype: width 2.8 mm; height 1.5 mm. Discussion. — All specimens at hand are crushed, except for the aperture and a portion of the ultimate whorl of paratype CM 41702, from which the distinctive flatly oval shape of the whorls at, and just behind the aperture is apparent. This specimen also shows an apertural dentical, and what may be the root of a broken- off, thin spiral denticle, about 0.5 mm long, on the right side of the penultimate whorl. This species generally resembles B. kishenehnsis (Russell, 1952), which Russell (1952:130,131) found at his W-1 and C-3 localities in British Columbia, and is very common at many localities of the Middle Fork Kishenehn Basin. Biomphalaria kishenehnsis is a much larger species, exceeding 35 mm with >6 whorls, bearing stronger costellae, and a heart-shaped cross section due to a mild peripheral angularity and only slight embracement of the preceding whorl. “Pla- norbis” cirrus White, 1877Z?, also bears a strong resemblance, but has more whorls, somewhat slower expansion of whorls, and a more rounded whorl section. Among modem species, B. spira closely resembles some forms of B. orbiculus (Morelet, 1849) in the flat oval cross section of the whorls, but is generally smaller, more depressed, and succeeding whorls of B. spira embrace preceding whorls to a greater degree. Environment and Distribution. — Richards (1937:255) found shells of B. orbi- culus abundant in dried swamps on the Island of Cozumel, Quintana Roo, Mexico. It is generally distributed along the Gulf Coastal region of Mexico, and on the Yucatan Peninsula (Baker, 1945). ‘"Planorbis” cirrus is known only from the middle Eocene (?Uintan-Bridgerian) of southwestern Wyoming (Henderson, 1935:245). Etymology. — Latin, spira, fern., coil, as of a serpent, or a rope, with reference to the flat, coiled appearance of this species. Material. — Five specimens. Holotype, CM 41702, is from South Ford Creek (NFF-22) locality; paratypes, CM 41703, KUMIP 289,762, KUMIP 289,763, and PPLN-123 are from Starvation Bend (NFF-12) locality (4), Flathead County, Montana, Lower Member, Kishenehn Formation, late Eocene- Oligocene. Subgenus Australorbis Pilsbry, 1934c Type species. — Planorbis guadalupensis Sowerby, 1822 {^P. glabratus Say, 1818). Environment and Distribution. — A tropical lake and river dweller, also found in marshes and bayous of South America and the Antilles. Biomphalaria glabrata (Say, 1818), a common modern species, is an extremely temperature sensitive species. Reproduction is limited to a water temperature range of between 20°C 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 73 and 30°C, and death occurs rapidly at temperatures of 5°C (Taylor, 1985:269). Fossil Australorbis are relatively common among Eocene faunas of the west, e.g., B. spectabilis (Meek, 1860Z?), B. utahensis (Meek, 1860Z?), and B. convolutus (Meek and Hayden, 1856). Biomphalaria kishenehnsis (Russell, 1952) (Fig. 16.D, 16.E) Planorbis kishenehnensis Russell, 1952:130-131; 1956:104-108. Australorbis pseudoammonius (Schlotheim, 1820). Taylor, in Ross, C. R, 1959:70-71; Taylor, 1975: 211. Biomphalaria kishenehnensis (Russell, 1952). Pierce, in Constenius et. al., 1989:197-198; Pierce, 1993:984. Discussion. — Russell (1952:130-131; 1956:104-106) reported this species from his localities W-1, E-3, C^l, C-3 (originally W-3 locality, his type locality), and an unnamed locality on the North Fork of the Flathead River in Montana just upstream of the mouth of Whale Creek (USGS 20198, Taylor, 1975:213). Spring floods and mass movements in the years since Russell made his collections have vastly altered the cut banks along the river. Of the localities listed above, W=l, C-1, and C-3 are destroyed or buried and the remnants of E-3 now contain a mostly terrestrial fauna. Only at Hook locality, across the river and slightly up- stream from Russell’s Whale Creek locality, but in a lithologically similar unit, have examples of B. kishenehnensis been found in the North Fork Basin. However, this species is commonly encountered at several, as yet undescribed, localities on the Middle Fork Basin (Fig. 16.D, from Coal Creek locality on Middle Fork). The material at hand, though limited, is adequate to confirm Russell’s Description and Remarks (1952:130-131; 1956:107-108), and his conclusion that B. kishe- nehnensis is closely related to, but distinct from, middle Eocene (Bridgerian) B. spectabilis (Meek, 18601?). Taylor {in McKenna et al., 1962) attempted a very broad revision of the fossil taxa within this genus. Basically, he decided that every large planorbid, “with numerous closely coiled whorls, nearly plane right side, concave left side, with a subangulation or abrupt curve next to the suture, and smooth nuclear whorls lacking carination” (Ibid., p. 8) constituted a single species, and that species was identical to the European ''Planorbina’' {Biomphalaria) pseudoammonius (Schlotheim, 1820). A diagnosis this broad is more suitable at subgenus or genus level than to define a species. Such diverse and readily separable taxa as Biom- phalaria convolutus (Meek and Hayden, 1856), B. spectabilis (Meek, 1860Z?), B. utahensis (Meek, 1860Z?) and B. kishenehnsis (Russell, 1952) would become sub- species, or, perhaps, simply local forms of a global species. I find this approach unacceptable, and reject this synonymization. Material. — Two specimens, CM 41704 and KUMIP 289,764, both fragmenary and incomplete, from Hook (NFF-26) locality, Flathead County, Montana, Lower Member, Kishenehn Formation, late Eo- cene. Figured specimen PPLN-125, of B. kishenehnsis, is from the Coal Creel locality in the Middle Fork part of Kishenehn Basin, some 22 km southeast of the Hook locality, where it is abundant, and is provided for comparison. Annals of Carnegie Museum VOL, 70 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 75 Tribe Planorbini Clarke, 1973 Genus Gyraulus Agassiz in Charpentier, 1837 Type species. — Planorbis albus Muller, 1774. Subgenus Torquis Dali, 1905 Type species. — Planorbis parvus Say, 1817. Environment and Distribution. — Gyraulids of the subspecies Torquis prefer shallow, quiet, well-vegetated ponds with almost any bottom, although mud seems to be preferred (Baker, 1928:377). Distribution is general in North America from Alaska to Sonora, Mexico, and Cuba (Bequaert and Miller, 1973:205). Previously known geologic range is uncertain, certainly since early Pliocene, possibly much older. Gyraulus, cf. G. scabiosus (Hanna, 1922) (Fig. 16.F-16.H) Planorbis (Torquis) scabiosus Hanna, 1922:4, 5, pi. 1, figs. 4—6. Discussion. — Hanna’s description and illustration of this species from an un- described bed of the Pliocene(?) Warner Lake beds of eastern Oregon matches almost exactly the specimens found in the Kishenehn Basin, especially in the shape of the aperture and the overlapping and abutting nature of the suture on the left side of these specimens. The description of the sculpture should be ex- panded to specify prosocline costellae on the right side that become radial to ophistocyrt after crossing the rounded perimeter of the whorls, and that the ap- erture is lunate-roundly ovate, advanced on the right side. There is no modem Gyraulus that has this exact combination of features. Measurements of figured hypotype, CM 41705: width 4.2 mm; height 1.2 mm; whorls —3.5, nucleus dam- aged. Material. — Nineteen specimens. Figured specimen, CM 41705, and KUMIP 289,765 are from Com- merce Creek (CMC-1) locality, British Columbia. Additional specimens are as follows: CM 41706 from E-5 (NFF-1) locality (1+2 fragments); PPLN-126 (2) from E-4 (NFF-2) locality, both British Columbia; and PPLN-127 (4) from Kintla Creek Carnegie (KTL-1) locality; KUMIP 289,766 (3) and PPLN-128 (4) from Wurtz Bend (NFF-1 9) locality; and PPLN-129 (3) from Starvation Bend (NFF- 12) locality, Flathead County, Montana, and all from Lower Member, Kishenehn Formation, late Eocene-Oligocene. Fig. 16. — BASAMMOTOPHORA: PLANORBIDAE. A-C: Biomphalaria spira n. sp. A-B: Holotype (CM 41702), X4; A. left; B. right; C. B. spira Paratype KUMIP 289763, X5, section through ultimate whorl behind aperture; D, E: Biomphalaria kishenehnsis (Russell, 1952); D. Hypotype PPLN-125, x2, left; E. Hypotype CM 41704, X1.5, interior of left side; F-H: Gyraulus sp., c.f. G. scabiosus (Hanna, 1922) Hypotype (CM 41705), X5; F. left; G. apertural; H. right. LYMNAEIDAE. I-J: Lymnaea bow- manana n. sp. Holotype (CM 41707), X5; I. apertural; J. lateral; K-L: Lymnaea newmarchi (Russell, 1952) Hypotype (CM 41709), X5; K. apertural; L. lateral; M-N: Lymnaea lacerta n. sp. Holotype (CM41711), X5; M. apertural; N. lateral. PHYSIDAE. O. Aplexa sp., c.f. A. hypnorum (Linnaeus, 1758) Hypotype (CM 41713), X5, apertural. BIVALVIA: SPHAERIIDAE. P-R:' Sphaerium progre- diens Russell, 1952; P-Q: Hypotype (CM 41715), X5; P. exterior of right valve; Q. umbonal of right valve; R. S. progrediens Hypotype (KUMIP 289774), X5, interior of left valve; S-U: Sphaerium discus n. sp., left valve Holotype (CM 41717), XIO; S. exterior; T. umbonal; U. interior. 76 Annals of Carnegie Museum VOL. 70 Family Physidae Fitzinger, 1833 Genus Aplexa Fleming, 1820 Type species. — Bulla hypnorum Linne, 1758. Aplexa sp., cf. A. hypnorum (Linne, 1758) (Fig. 16.0) Discussion. — Although a rarely encountered species in the Kishenehn localities, and despite not having at hand a “perfect” specimen, I have no hesitation in referring these specimens conchologically to the genus Aplexa, and they compare favorably with modern A. hypnorum. I have compared these specimens with sev- eral lots of late Pliocene to modem A. hypnorum specimens from High Plains localities that are in my reference collection. Other than being slightly more slen- der, on an average, no significant difference can be discerned. This is yet another of the taxa in this Kishenehn fauna the appears to have survived unchanged conchologically from ?late Eocene to the Recent. As elsewhere stated, this does not imply biologic specific identity. Of the four specimens at hand, two are ju- veniles, heights 2.6 mm and 3.3 mm, with 2.5 to 2.75 whorls. The largest spec- imen, from Bowman Creek locality, has both apical and basal breakage, but is estimated at a height of 7 mm with 3.5 to 4 whorls. The previously known geologic range of this species is from late Pliocene, Blancan (Pierce, unpublished data) to modern. However, this is probably the same as Aplexa indet. reported by C. P. Ross (1959:70-71) from his Locality 14778, which is very near our Bowman Creek (BWN-2) locality. Among fossil sinistral taxa, two Cretaceous species, Bulinus atavus White, ISlla, now considered an Aplexa, and B. disjuncta White, 1877Z?, now considered a sinistral Lymnaea (Pierce, 1996), are most similar. Both are much larger than A. hypnorum. Although Clarke (1973:383) reports height up to % inch for A. hypnorum in Canada, fossil and modern shells from the High Plains seldom exceed 7 mm in height, which is the estimated height of the largest specimen from Bowman Creek locality. In addition to size, Aplexa atava is a considerably more elongate species, with a W/H of only 0.33, compared with 0.40 to 0.45 for A. hypnorum. Lymnaea disjuncta can be distinguished by its inflated body whorl and prominent columellar plication. Russell (1956:104-108), reported a Physal from his C-3 locality in British Columbia, and from an unnamed locality on the west bank of the Flathead River near the mouth of Whale Creek. His restoration is of a much larger physid, more similar to some of the slender morphs of Physa gyrina Say, 1821. A physid of this character was not encountered as part of this study. Environment and Distribution. — This is a species of seasonal streams and pools, often with mud bottom, occasionally from small, clean brooks with mud bottom. Distribution is holarctic, in North America from the Arctic Ocean south to Colorado and Utah in the west and to about the Ohio River in the east, and from Washington east of the Cascades east to the Atlantic Ocean (Baker, 1928: 474; Taylor, 1966:110). Geologic range of the genus is unclear, since the true taxonomic position of many late Mesozoic and early Cenozoic species variously assigned to the genera '' Bulinus Physa, mvd Aplexa is unresolved. It may extend from Cretaceous to Recent; for A. hypnorum it is certainly from late Pliocene to Recent. Material. — Four specimens. Figured specimen CM 41713 is from Bowman Creek (BWN-2) locality, Flathead County, Montana. Additional specimens are as follows: KUMIP 289,773 from E-4 (NFF-2) locality, British Columbia; and CM 41714 (2) from Starvation Bend (NFF-12) locality, Flathead 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 77 County, Montana. All specimens from the Lower Member, Kishenehn Formation, late Eocene-Oli- gocene. Superfamily Lymnaeoidea Family Lymnaeidae Rafinesque, 1815 Genus Lymnaea Lamarck, 1799 Type species. — Helix stagnalis Linne, 1758. Discussion. — The lymnaeids are notorious for having highly variable shell forms (ecophenotypes) within the confines of an single anatomical species. As a result, there is a plethora of “species” based on variations of shell form. Huben- dick (1951), based on anatomical studies, recognized only the genera Lanx Cles- sin, 1882, and Lymnaea in North and Central America, with only 13 species (two questionable) within the latter, and illustrated the known range of shell variability for each species. The species recognized by Hubendick closely approximate the “groups” as used by Baker (1911, e.g., p. 298). For paleontological usage, since anatomical character for determination of species is not possible, defining “spe- cies” as members of “groups” based on the defined range of shell variability for a modem species (sensu Hubendick, 1951) appears to be the most workable so- lution (Pierce, 1993:984-985). This usage appears to have no effect on environ- mental or ecological considerations. Group of Lymnaea palustris (Muller, 1774) Discussion. — The shell of this group is one of the most clear-cut and easily defined of the lymnaeids (see Hubendick, 1951:119-120 for diagnosis.) Environment and Distribution. — Prefers lakes of any size, well vegetated, clear, or stagnant. Malleated forms are typical of stagnant water and muddy bottoms (Baker, 1911:331). Distribution holarctic, in North America, south to California, New Mexico, and Missouri, thence northeast to Nova Scotia (Hubendick, 1951: fig. 304). The genus Lymnaea is ancient, extending back at least as far as Jurassic. Since there has been no comprehensive study of fossil lymnaeids, the range of the various Groups is not clear. The Group of L. palustris extends back to Eocene (Bridgerian), at least with L. vetusta Meek, 1860Z?. Lymnaea bowmanana Pierce, new species (Fig. 16.1, 16.J) Diagnosis. — Shell elongate conic, of six or more moderately rounded whorls, spire high, acute, aperture elongate ovate-lunate, columellar and basal lips ex- panded. Description. — Shell dextral, of moderate size, elongate conic, spire high, acute, <45°, whorls six or more, moderately rounded, slightly shouldered, rapidly elongating, suture moderately impressed, occasionally appearing flanged and/or crenulated, nucleus small, diameter —0.25 mm, nuclear whorls about 1.5, smooth and shining, subsequent whorls with fine growth costellate, sharpest near suture, blurring to wrinkles below periphery of whorl, last whorl often malleated, narrowly umbilicate. Ap- erture elongate, Ha/H > 0.40, narrowly ovate-lunate, outer lip thin, sharp, simple, basal lip widely but shallowly reflected, columellar lip expanded, rolled over and obscuring umbilicus, columella with prominent plication just below parietal margin. Measurements of holotype: height, 11.3 mm; width 4.7 mm; height of aperture, 4.8 mm; W/H 0.40; Ha/H 0.41; whorls 6.1. Largest specimen observed is 18.5 mm in height from South Ford Creek locality. Discussion. — This species is one of the most widely distributed lymnaeids in the Kishenehn basin. It resembles, of course, all other members of the group of 78 Annals of Carnegie Museum VOL. 70 Lymnaea palustris. Among fossil members of this group, L. bowmanana most resembles Eocene L. vetusta Meek, 1860Z?, and Oligocene L. ambiguspira Pierce, 1996, both in the group of L. palustris (Pierce, 1993:985-986). Lymnaea bow- manana can be easily distinguished from them by its shorter, narrowly ovate- lunate aperture, and flared, spoon-shaped basal lip. Among modern “species” of the L. palustris Group, it most resembles L. elrodiana Baker, 1935, which has a more widely ovate-lunate aperture and less prominently flared basal lip. It is most interesting to note that L. elrodiana is living in nearby Lake McDonald (Elrod, 1902), part of the Flathead River drainage system. Etymology. — bowmanana, with reference to the type locality on Bowman Creek, a tributary of the North Fork of the Flathead River, Montana. Material. — More than 140 specimens. Holotype CM 41707 from Bowman Creek (BWN-2) locality, Flathead County, Montana, Lower Member, Kishenehn Formation, late Eocene-Oligocene. Paratypes: KUMIP 289,768 (8) and PLYM-116 (13), from Bowman Creek (BWN-2) locality; KUMIP 289767 and PLYM-1 17 (6) are from South Ford Creek (NFF-22) locality, CM 41708 (8) and PLYM-1 18 (50) are from Wurtz Bend (NFF-19) locality, all Flathead County, Montana; PLYM-1 19 from Moose City North (NFF-9) locality; PLYM-1 20 from W-2 (NFF-5) locality; PLYM-121 from E-2 (NFF-6) locality; and PLYM-1 22 (>50) from E-4 (NFF-2) locality, all British Columbia. All localities in Lower Mem- ber, Kishenehn Formation, late Eocene-Oligocene. Group of Lymnaea catascopium Say, 1817 Discussion. — This group is highly variable in shell form, but is generally dis- tinguished by an “aperture ovate or long ovate, somewhat produced and effuse anteriorly, as long as, or longer than the spire; . . . inner lip narrow, reflected and appressed tightly to the columellar region, either completely closing the umbilicus or leaving a very small, narrow chink; columella with a heavy oblique plait caus- ing the axis to be twisted; ...” (Baker, 1911:382). In this group, the fusiform shape and elongated aperture are key traits. Environment and Distribution. — Characteristic of larger lakes, rivers and bays, often near shore, on most bottoms, but seems to prefer a rocky shoreline (Baker, 191 1:387). A North American group with main distribution in north central United States and south central Canada, with outlying species from British Columbia to California and east to Utah (Hubendick, 1951:fig. 315). Geologic range is uncer- tain. Lymnaea newmarchi Russell, 1952 (Fig. 16.K, 16.L) Emended description. — “Shell small, slender, narrowly elongate conic, spire high and acute, spire angle <40° with about five whorls, nucleus small, --0.30 mm, nuclear whorls about 1.5, smooth to granular, subsequent whorls becoming fiatly rounded, suture shallow, the body whorl rather elongate. Aperture narrowly ovoid, the length exceeding that of the spire, Ha/H > 0.50, narrowly and acutely angled at apex, parietal callus continuous, inner lip narrow but thickened, a solid, sinuous extension of columella, completely obscuring umbilicus, with a slight plication just in front of the parietal portion, basal lip weakly hut broadly reflected, elongating aperture. Surface marked by fine, rather regular growth lines, strongest near suture, variable on remainder of whorl, ranging from continuing undi- minished onto basal portion of whorl to becoming indistinct fine growth wrinkles, strongest on body whorl beginning below sutural flange; color white to pale bluish gray, but some specimens with pink coloring, which marks the growth lines and also appears as fine revolving lines” (after Russell, 1952: 103, emmendations in italics). Measurement of hypotype: height E9.6 mm; width 3.3 mm; height aperture E5.6 mm; W/H 0.34; Ha/H 0.58; whorls 4.7, outer and basal lip of aperture missing. Largest specimen observed, 18.0 mm, crushed, is from Starvation Bend locality. Discussion. — Russell’s (1952:129-130) original description was based on an 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 79 immature specimen of only four whorls, although he had a larger specimen (h=lL0 mm). Russell’s type locality, apparently on a displaced slide block, has been eroded away. Lymnaea newmarchi appears to be of the Lymnaea catasco- pium group, although resembling several of the more slender morphs of the mod- em L. palustris group, such as L. elodes Say, 1821 and L. exilis Lea, 1834. Although the spires of both are very similar to L. newmarchi, their apertures are much shorter, and, comparatively, more inflated, and both fall short of the exag- gerated reflected flare of the basal lip that so elongates the aperture of L. new- marchi. Most similar is L. sumassi Baird, 1863, from British Columbia and Utah, whose taxonomic position is open to some question, e.g., Hubendick (1951:205) considers it a L. palustris, whereas Baker (1911:XV, 403) lists it as a member of his Group of Catascopium. Considering Hubendick (1951:130) adopted Baker’s (1911:382) description of L. catascopium, and the apertural characteristics tend to support Baker’s choice, it is adopted herein. Lymnaea sumassi, especially the type specimens illustrated by Baker (1911:pl. 41, figs. 11-14) differ only in being much larger and having a somewhat shorter, more inflated aperture. Among the older fossil lymnaeids with characteristics similar to L. newmarchi, only Eocene L. vetusta and Oligocene L. ambiguspira are close enough for comparison. Lym- naea newmarchi can be easily differentiated by its more slender shape, flatly rounded whorls, and slender, elongate aperture. Pleistocene L. nashotahensis Bak- er, 1909, from Wisconsin, considered a L. palustris by Hubendick (1951:196), is remarkably similar, but has more rounded whorls, a shorter, inflated aperture and “ ... an umbilical chink which is usually absent in catascopium"' (Baker, 1911: 109). In short, the exaggerated character of the aperture of L. newmarchi appears unique among the lymnaeids. Material. — More than 30 specimens. Figured specimen, CM 41709, and KUMIP 289,770 (2) are from Wurtz Bend (NFF-19) locality, Flathead County, Montana. Additional specimens: CM 41710 (3) and PLYM-123 (12 cr.) are from Starvation Bend (NFF-12) locality, Flathead County, Montana; KUMIP 289,769, PLYMU24 (1 1) are from E-5 (NFF-1) locality (11); and PLYM-125 from Commerce Creek (CMC-1) locality (2); and PLYM-132 (2) from W-2 (NFF-5) locality, all British Columbia. All localities in Lower Member, Kishenehn Formation, late Eocene-Oligocene. Group of Lymnaea bulimoides (Lea, 1841) Discussion. — It is very difficult to consistently separate, by shell only, the many similar morphs found among the groups of L. bulimoides, L. humilis (Say, 1822), and L. cubensis Pfeiffer, 1839. Environment and Distribution. — Apparently a species of ephemeral ponds, roadside ditches, small, slow streams, usually well vegetated if permanent, bot- toms often muddy (Baker, 1911:216, 220; Franzen and Leonard, 1943:405; Clarke, 1973:286-287). Distribution is generally west of the Mississippi River, from British Columbia south to the approximate U. S. / Mexican border, east to Kansas and Louisiana (Hubendick, 1951:fig. 313). In the far west, characteristic of the Columbian and Californian Life Zones (Baker, 1911:211). Geologic range is uncertain, but is previously known at least from late Pliocene (Blancan). Lymnaea lacerta Pierce, new species (Fig. 16.M, 16.N) Diagnosis. — A small, thin, umbilicate shell with shouldered whorls, a domi- nant, rugose body whorl and narrowly oval aperture. 80 Annals of Carnegie Museum VOL. 70 Description. — Shell small, thin, of about five regularly and rapidly expanding whorls, body whorl dominant, suture impressed, spire subacute but rather bulbous, nucleus small, —0.25 mm, nuclear whorls fewer than two, smooth and rounded, subsequent whorls modestly shouldered, with very fine, regularly spaced, near orthocline costellae, ultimate whorl bears broad and low rugae with costellae superimposed, occasionally appearing malleated; aperture narrowly oval, Ha/H —0.50, outer lip thin, simple, peristome continuous, columellar lip sinuous, weakly plicate, slightly thickened, broadly ex- panded and folded over, but not hiding, the umbilicus, basal lip joins the columellar lip with a roundly acute angle. Measurements of holotype, an immature specimen: height 4.6 mm; width 2.8 mm; HAV 1.71; Ha 2.25 mm; Ha/H 0.49; whorls 4.25. Largest specimen observed, height >9.0 mm, whorls >5, badly crushed, is from South Ford Creek locality. Discussion. — Lymnaea lacerta is a rather rare component of the lymnaeid fauna of the Kishenehn Basin, being found in small numbers at six localities, and, with the rare exception of the Bowman Creek locality, the thin shell is almost invari- ably crushed. Among modem lymnaeids, L. lacerta resembles L. cubensis, several morphotypes of L. humilis, and, most closely, the techella, Haldeman, 1867, morph of L. bulimoides. The latter can be distinguished from L. lacerta by its obviously malleated body whorl, lacking rugae, and longer, wider aperture. Fossil lymnaeids of the bulimoides/humilis groups, although rather common after the mid-Pleistocene (Kansan) in faunas from the High Plains, are but rarely encoun- tered in strata as old as late Blancan. Older records are, apparently, unknown. The only taxon of similar age that might be confused with L. lacerta is Oligocene L. tumere Pierce, 1993, from the Cabbage Patch fauna of southwest Montana. Lymnaea tumere belongs to the group of L. stagnalis Linne, 1758, and can be easily identified by its low, pinched spire, and inflated D-shaped aperture. Etymology . — Latin, lacerta, fern., brawny, with reference to robust outline of this species. Material. — More than 35 specimens. Holotype: CM 41711 from Bowman Creek (BWN-2) locality, Flathead County, Montana, Coal Creek Member, Kishenehn Formation, middle Oligocene. Additional specimens: KUMIP 289,772 (7) and PLYM-126 from Bowman Creek (BWN-2) locality; PLYM-128 from South Ford Creek (NFF-22) locality; PLYM-127 (15) from Wurtz Bend (NFF-19) locality, all Flathead County, Montana; KUMIP 289,771 and PLYM-129 from Moose City North (NFF-9) locality; CM 41712 (5) and PLYM-130 from W-2 (NFF-5) locality; and PLYM-131 (5) from E-4 (NFF-2) locality, all British Columbia. All localities in Lower Member, Kishenehn Formation, late Eocene- Oligocene. Class Bivalvia Linne, 1758 Order Veneroidea H. and A. Adams, 1856 Family Sphaeridae Dali, 1895 Subfamily Sphaerinae Baker, 1927 Genus Sphaerium Scopoli, 1777 Type species. — Tellina cornea Linne, 1758. Subgenus Sphaerium s. str. Sphaerium progrediens Russell, 1952 (Fig. 16.P-16.R) Description. — “Shell of moderate size for the genus, almost symmetrical in side view. Beak low on smaller shells, and placed near midlength; in larger shells more prominent, with a forward incli- nation. Dorsal margins short, sloping from beak; extremities broadly rounded except posterioventral portion, which may be narrowly rounded in less symmetrical examples; ventral margin gently convex. Surface marked by fine, irregular growth lines and on some shells by a growth furrow. Dentition of right valve: one distinct, oblique cardinal tooth, and a toothlike ridge at base of hinge plate; anterior and posterior lateral socket, each with toothlike ridge above and below. Dentition of left valve: two oblique cardinal teeth, separated by a distinct socket, the lower or more anterior tooth having its socket curved upward and backward slightly to constrict opening of socket; one anterior and one posterior lateral tooth, each rising from ventral margin of hinge plate, and each bounded above by a broad longitudinal groove; anterior lateral tooth more prominent, almost pointed” (Russell, 1952:129). 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 81 Emmendation. — Valves sub-circular in outline, modestly inflated, walls moderately thick, sculpted with prominent, widely but irregularly spaced, growth furrows (striae), giving appearance of banding, umbonal area smooth, remainder of valve with very fine, closely and more regularly spaced striae between the growth furrows, hinge moderately long and wide, wider at cusps, cusps robust, rounded to slightly angular at top, PHI has a distinct, longitudinal furrow, C3 short, considerably curved, posterior end much heavier. Measurements of hypotype, CM 41715, a right valve: width 9 mm; height 8 mm; thickness 2 mm. Discussion. — Among modem sphaeriids, S. progrediens resembles, in general shape and sculpture, S. occidentale Prime, 1860, but differs markedly in strength of sculpture. The growth furrows of S. progrediens are stronger, more numerous, and more closely spaced, resulting in a distinctly banded appearance. Among fossil sphaeriids, S. progrediens is distinct in shape and in sculpture, and unlikely to be confused with any other except Eocene S. formosum (Meek and Hayden, 1856), which, although similar in shape, has much less robust laterals. Russell’s (1952:129) type locality for this species, and for the two unionid taxa he reported from the North Fork of the Flathead River, has been destroyed by erosion. Environment and Distribution. — Subgenus Sphaerium (s. str.) is widely dis= tributed across Canada and northern United States. Normal habitat is creeks, rivers and small lakes on mud or sandy mud bottom, usually well vegetated (Herrington, 1962). Geologic range from Upper Jurassic (?) or Lower Cretaceous to Recent (Clarke, 1973:135). Material. — More than 50 specimens. Figured specimens: CM 41715 from E-4 (NFF-2) locality, and KUMIP 289,774 from W-2 (NFF-5) locality, are both British Columbia. Additional specimens: PSPH- 101 (9) from E-5 (NFF-1) locality; PSPH-102 (10) from E-4 (NFF-2) locality; CM 47176 (15) from W-2 (NFF-5) locality, all British Columbia. KUMIP 289,775 (15) from Starvation Bend (NFF-1 2) locality; PSPH-103 (3) from Wurtz Bend (NFF-19) locality; and PSPH-104 (8) from Bowman Creek (BWN-2) locality, all Flathead County, Montana. All localities in the Lower Member, or the Coal Creek Bed of the Basal Member, Kishenehn Formation, late Eocene-Oligocene. Subgenus Musculiurn Link, 1807 Type species. — Tellina lacustris (Muller, 1774), by subsequent designation (Morsch, 1 862:228, Clark, 1973:151). Sphaerium discus Pierce, new species (Fig. 16.S-16.U) Diagnosis. — A small Sphaerium with circular valves, and slightly protruding, weakly calycuate umbones. Description. — Valves small, circular in outline, inflated, thin, sculpted with fine, closely spaced striae that fade on umbonal region, becoming coarser, stronger, and more widespread as ventral margin is approached, beaks central, protruding, elevated, slightly calyculate, dorsal margin short, sloping sharply from beak, anterior, ventral and posterior margins well rounded, near circular. Lateral teeth are somewhat small and weak. Measurements of holotype, CM 41717, a single right valve: width 3.9 mm; height 4.0 mm; thickness 1.1 mm. Discussion. — This specimen is so unique that it is difficult to find a species, fossil or living, with which to compare. The produced, slightly calyculate beaks are similar to modern and fossil S. partumeium (Say, 1822), common in late Blancan faunas of Nebraska. However, its circular valve outline immediately dis= tinguishes this species from all other sphaerids. Etymology. — discus — Latin, with reference to the shape of the valve. Environment and Distribution. — The subgenus Musculiurn is found from Sub- 82 Annals of Carnegie Museum VOL. 70 Arctic Canada and Alaska south through the United States, Mexico, Central Amer- ica into South America (Burch, 1972), and, locally, in ponds near Flathead Lake, Montana (Elrod, 1902:103). It inhabits almost all types of permanent water, from ponds and lakes to streams with rapid current, generally on mud bottom with some to abundant vegetation, and can be found in apparently seasonal pools. Previously known geologic range is from Miocene to Recent (Clarke, 1973:151- 162). Material. — Five specimens. Holotype, CM 41717, a right valve, and paratype KUMIP 289,777, a partial valve, are from Starvation Bend (NFF-12) locality. Paratypes KUMIP-289,776, a slightly com- pressed left valve, and CM 41718, a valve fragment, are from South Ford Creek (NFF-22) locality, both localities are in Flathead County, Montana. PSPH-105, a very juvenile valve, tentatively referred to this species, is from E-4 (NFF-2) locality, British Columbia. All localities in Lower Member, Kishenehn Formation, late Eocene-Oligocene. Discussion Russell's Fauna. — Surprisingly, despite a rich and diverse fauna recovered, and the depth of this investigation, not all the taxa described by Russell (1952; 1956) were found. Those not found are as follows: Taxa Russell’s Locality Triodopsis buttsi Russell, 1956* E-4 Holospira adventica Russell, 1956 C-2 Ventridensl sp. C-2 Zonitoides? sp.** C-2 Physa sp. C-3 Eliptio salassiensis Russell, 1952 W-1 Lampsilis dayli Russell, 1952 W-1 * — A smaller subspecies was found nearby. ** — May be an incomplete Polygyroides montivaga Pierce, 1992. Nearly all fossil localities investigated by Russell, and during this investigation, are at, or only slightly above, normal stream flow levels for the North Fork. As such, these fossil localities are subject to severe scour by fast, turbulent flow during the strong spring floods, and subsequent slumping of overlying strata, often entirely eroding or covering the exposed fossiliferous bed over a period of a few years. Even within the short time limits of this investigation, major damage to fossil localities are noted, on a year to year basis. From Russell’s (1952; 1956) descriptions and photographs, it is apparent that some of his localities now appear very different than when Russell made his collections. Some localities could not be found. For example, W-1 is, apparently, completely destroyed by erosion, and the Couldrey Creek localities (C-2, C-3) are covered by slumping and heavy vegetation. Taxa unique to these localities were not rediscovered. Although our collections from the localities for which we deign to use Russell’s designations are probably from within only a few meters of Russell’s actual localities, it must be understood that they may not include exactly the same strata collected from by Russell. For example, our collections from the locality referred to as E-4 (NFF- 2) comprised 20 taxa, compared to only five taxa collected by Russell (1956:105) from his locality E-4. Even so, our collection includes only two of the taxa he reported from his E-4 locality. Of the remaining three taxa, ‘'Triodopsis” buttsi was found elsewhere, but only as a much smaller subspecies, “Binneya” antiqua 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 83 was found at adjacent localities upstream, and Holospira adventica, which Russell reported here based on fragments only, has not been rediscovered, although frag- ments of another urocoptid, Coelostemma dawsonae, were found at this locality. The large amount of material collected as part of this investigation permitted correction of some of the taxonomy used by Russell. A list of changes follows: Russell’s taxonomy New taxonomy Triodopsis butt si Binneya antiqua Anguispira simplex Stagnicola newmarchi Planorbis kishenehnsis Gyraulus procerus Goniobasis sp. = Tozerpina buttsi (Russell, 1956) Ceresidae = Omalonyx antiqua (Russell, 1956) Succineidae = Haplotrema simplex (Russell, 1956) Haplotrematidae = Lymnaea newmarchi (Russell, 1952) Lymnaeidae = Biomphalaria kishenehnsis (Russell, 1952) Planorbidae = Valvata procera (Russell, 1952) Valvatidae = Tryonia russelli Pierce, n.s. Hydrobiidae Living species. — Among the surprises encountered in the Kishenehn l.f. were four taxa that could not be distinguished, conchologically, from living taxa. These taxa are as follows: Discidae Ammonitellidae Planorbidae Physidae Anguispira sp., cf. A. alternata (Say, 1816) Polygyrella sp., cf. P. polygyrella (Bland and Cooper, 1861) Planorbula sp., cf. P. campestris (Dawson, 1875) Aplexa sp., cf. A. hypnorum (Linne, 1758) These are not just represented by fragments or by single, deformed shells. The rarest of those above is Aplexa sp., cf. A. hypnorum, had four good specimens at hand, whereas Anguispira sp,, cf. A. alternata numbered more than 60 specimens, including one very nice specimen lodged in, and cemented into, the aperture of the holotype of Skinnerelix rothi (CM-41676) (Fig. 13.B). Despite the fact that the preserved conchological characteristics of these four taxa fall well within the expected range of variation for the modem species, actual biological identity is considered unlikely, and is not implied. However, the habitat preferences and distribution of the modern species were used in studies of sympatry. Sympatry. — Sympatry is defined, simply, as the natural coexistence of a group of taxa in a given area at a given time. Since climates and environments change with time, the study of the taxa of an ancient fauna or flora can provide valuable clues as to the climate and environment in which they existed. Mollusks, as a result of their low degree of mobility and apparent slow rate of evolutionary change, are excellent indicators of the climate and environment in which they exist. Terrestrial gastropods, especially, closely reflect their environment; in effect, their low mobility means that they are ‘stuck with the hand they were dealt.’ Unlike mammals, who are highly mobile and migratory, terrestrial gastropods are unable to escape even slow climatic changes. If the climate changes to one beyond their tolerance range, local extinction results. Freshwater mollusks, however, by virtue of their environment, are shielded, at least in part, from the outside terres- trial environment, and are able to survive in isolated suitable microenvironments. For example, cool springs in northern Nebraska harbor taxa adapted to Canadian environments (Taylor, 1960). They are relicts of the Pleistocene. The rich, diversified, and, at least locally abundant, molluscan fauna of the North Fork Kishenehn Basin clearly indicates a paleoclimate at considerable odds 84 Annals of Carnegie Museum VOL. 70 with the present climate. Since most taxa of the Kishenehn l.f. are extinct, to determine plausible climatic and environmental conditions for the time of depo- sition of the Kishenehn strata, it was necessary to identify modern analogs of the fossil taxa. It is assumed that the environmental and climatic preferences of these analogous modern taxa will approximate that of the fossil taxa. With these modern analogs, modern areas of sympatry can be established. The climate and environ- ment of these modem areas of sympatry are then used to infer the climate and environment at the time of deposition. In the case of those four species that are indistinguishable from modem taxa, a modern analog was unnecessary. Several taxa from among the Planorbidae, the Hydrobiidae, and Valvatidae also resembled modem species so closely that the modern species were used as analogs. In some cases, such as Vallonia kootena- yorum, clearly a member of the living V. cyclophorella group, the choice was, again clear. For Gastrocopta minuscula, which resembles modern G. pellucida (s. lat.), and since both are of the same subgenus, the modem species was selected as the stand-in. For a large number of the remaining taxa, the Pupillidae, Planor- bidae, and Urocoptidae, and some Helicinidae, those new taxa assignable to mod- em subgenera were assumed to have the environmental and climatic preferences of the modem subgenus as a whole. In other cases, it was a composite of the requirements of the modern genus that was the best possible match, e.g., Menetus hilli, the Oreohelicidae, and the Succineidae. The analogs of the Lymnaeidae were determined by their Group characteristics (see discussion on Groups under Sys- tematics, this paper). In extreme cases, as with the Helicinidae and Ceresidae, it was the composite character of the subfamily or even of the family as a whole that was deemed as the best possible modern analog. Two taxa, Vertigo doliara, and V. micra, were deemed unsuitable for use in this exercise, since they could not be classified to sub-generic level, and the modem distribution at generic level was too broad to be significant, and the aquatic genus Tryonia, imperfectly known, and extinct, was also excluded. Tables 4 and 5 show the analogous taxon selected for each Kishenehn taxon, and the current range of each. The modem ranges of the analog taxa were mapped. No single area of sympatry for the North Kishenehn Basin l.f., as a whole, was discovered. In actuality, three more or less overlapping areas of sympatry developed: Group I — Tropical Wet environment; Group II — Subtropical semiarid environment; and Group III — Tem- perate environment. Figures 17 and 18 define the areas of sympatry for Groups I and II. Modem climatic parameters for each of the sympatric areas, as well as for the North Kishenehn Basin, were determined using climatic data available from www.worldclimate.com on the World Wide Web. Although the taxa of the Kishenehn l.f. comprise a broad melange overall with respect to interpreted habitat and environment preferences, they separate naturally into these three climatic groups. In Table 6, the various terrestrial taxa of the Kdshenehn fauna are separated into these Groups, with each Group representing a distinct climate preference. Group I. Tropical, Wet and Dry, Savannah, Aw- As (Koppen, 1931). This Group is comprised of seven terrestrial taxa, five of which have modern analogs restricted to this climate. This Group covers a large area, displaced 25°-40° in latitude South and 15°-55° in longitude East from the Kishenehn Basin. Although the climatic and environmental needs and preferences of the seven members of this group are quite constant, they occupy widely dispersed geographic areas with only G. pel- lucida (s. lat.) distributed throughout the total area. Only two of the modem analogs, G. pellucida (5. lat.) and Praticolella, range into other climate groups. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 85 Table 4. — Analogous terrestrial taxa. terrestrial taxa analogous taxa CURRENT RANGE CERESIDAE Tozerpina buttsi parva E Ceresidae Eastern Mexico T. lends E Ceresidae HELICINIDAE Waldemaria monticola Hendersonia WI, lA, southern Appala- chians Lucidella salishana Lucidella! Poeniella Greater Antilles, Florida Keys, L. Columbiana Lucidella! Poeniella Cent. America & Gulf Coast of Mexico SUCCINEIDAE Oxylomal kintlana O. ejfusa Gp. East of Continental Divide Omalonyx cocleare Omalonyx S. & C. America, Lesser An- tilles PUPILLIDAE Pupoides costatus Ischnopupoides Four Corners area, Black P. tephroides Isclmopupoides Hills, & Front Range, CO to NM Gastrocopta miniscula G. pellucida Carribbean, Gulf Coast, W. to Baja CA G. kintlana Albinula East of Continental Divide G. akokala Albinula G. leonardi Vertigopsis East of ContT Divide, UT- Guatemala Vertigo consteniusi Vertigo s.s. Widespread, Canada, U.S., Vertigo, c.f V. arenula Vertigo s.s. Mex. VALLONIIDAE Vallonia kootenayorum V. cyclophorella Gp. AZ, NM, West TX, Sonora, Chihuahua HAPLOTREMATIDAE Haplotrema simplex Haplotrema Coast Ranges of So. CA & Baja CA UROCOPTIDAE Holospira tabrumi Haplocion West Texas, Coahuila, Chihua- hua H. beardi Haplocion Coelostemma dawsonae Coelostemma Guerrero, Oaxaca, Coahuila, N. Leon DISCIDAE Discus mackenziei Discus N. U.S. and S. Canada Anguispira, cf A. alternata Anguispira alternata E of 97th Meridian & Colum- bia River LIMACIDAE Deroceras mahiz D. laeve Alaska to Central America D. securis D. laeve ZONITIDAE Nesovitrea pulchra Nesovitrea hammonis 37th parallel N to Arctic Ocean POLYGYRIDAE Praticolella lucifera P. berlandieriana Gulf Coast from TX to Yuca- tan Ashmunella sp. Ashmunella AZ, NM, West TX, & Chi- huahua 86 Annals of Carnegie Museum VOL. 70 Table 4. — Continued. TERRESTRIAL TAXA ANALOGOUS TAXA CURRENT RANGE OREOHELICIDAE Oreohelix dawsonae Oreohelix S. BC to S. NM Radiocentriim kintlana Radiocentrum S. NM, Chihuahua & Baja CA AMMONITELLIDAE Polygyroidea montivaga Polygyroidea Cent. CA Polygyrella, cf. P. polygyrella P. polygyrella OR, WA, ID, MT HUMBOLDTIANIDAE Skinnerelix rothi Hiimboldtiana W. TX thru Cent. Mexico to D.F. HELMINTHOGLYPTIDAE Xerionata constenii Xerarionta Coastal CA and Baja Califor- nia Table 5. — Analogous aquatic taxa. AQUATIC TAXA ANALOGOUS TAXA CURRENT RANGE HYDROBIIDAE Cincinnatial howmanana C. cincinnatiensis ND, UT, Ohio drainage Fluminicolal calderen se F. fiisca OR VALVATIDAE Valvata procera V. humeralis BC, MT, CA, NM, & C. Mex V. procera spatiosa V. humeralis PLANORBIDAE Helisoma triangulata Helisoma anceps Canada, U.S., Mexico Planorhella fordensis P. subcrenulatum YT, MB, MI, CO, CA Planorhula, c.f. P. campestris P. campestris BC, MB, NM Biomphalaria hay deni Tropicorbis Gulf States, Mexico, Caribbean B. spira Tropicorbis B. kishenelmsis Aiistralorbis S. & C. America, Lesser Antilles Menetus textus Menetus cooperi CA, AB, AK M. (M.) hi Hi s. gen. Micromenetus East of Mississippi R, and lA to TX Gyrauliis, c.f. G. scabiosus s. gen. Torquis AK to Sonora to Cuba LYMNAEIDAE Lymnaea bowmanana L. palustris Gp. Holarctic to CA, NM, MO L. lacerta L. bulimoides Gp. W of Miss. R., BC to AZ & NM L. newrnarchi L. catascopium Gp. Cent. U.S. & Canada & BC to CA PHYSIDAE Aplexa, c.f. A. hypnorurn A. hypnorurn Holarctic to WA, CO, OH, MD BIVALVES SPHAERIDAE Sphaerium progrediens Sphaerium Canada & U.S. north of 37°N S. (M.) discus s. gen. Musculium Canada & U.S. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 87 Figure 17. — Group I Area of Sympatry. Gastrocopta pellucida (s.l.) salient. The multiple sub-centers of sympatry for these taxa are as follows: The Antilles (Caribbean) (4 taxa); eastern San Luis Potosi, Mexico (3 taxa); Yucatan Peninsula (2 taxa); and northern South America (2 taxa). In common, these taxa require, or at least tolerate, a very constant tropical climate: MAT 25°-27°, Annual Range 3°“10°C, Annual Precipitation >100 cm, and seasonally winter dry (www.worldclimate.com), and are generally found at relatively low elevations. Group IL Subtropical/Tropical Semiari^Arid, Bsh~Bwh, Cs. The composite ranges of the 17 terrestrial taxa in this group cover a broad area the Southwestern United States and Mexico, from the Gulf Coast to the Colorado River and Baja California, with three distinct sub-centers. However, climatic data determined that the three /oci. East, Central and West, had enough climatic parameters in common to be treated broadly as a single sympatric area, despite the rather large longitu- dinal separation between the individual centers. These taxa are displaced 16°-33° in latitude South and in longitude from straight south to 7°--14° East. In Table 6, Group II is shown split into its sections. An Eastern section, HE, composed of 5 taxa, is centered in eastern Coahuila and Nuevo Leon. Eor this section, HE, the climate and environment of Monterrey, Nuevo Leon, Mexico is a good approxi- mation: MAT 22°C, Annual Range 13.5°C, Annual Precipitation 61 cm, and sea- sonally winter dry, with terrain elevations of —500 m. A Central section, IIC, composed of 12 taxa, centers in the lower Rio Grande Valley of New Mexico, and adjacent Chihuahua, and is generally representative of the Southwestern Mol- 88 Annals of Carnegie Museum VOL. 70 Figure 18. — Group II Area of Sympatry. Gastrocopta pellucida (s.l.) salient ■■■ ■■•■Praticolella herlandehana salient -•-■-■-Ischnopupoides salient luscan Province (sensu Henderson, 1931:182). For this Central section, IIC, the suggested climate and environment would be an average of that occurring at Elephant Butte Reservoir, NM and Winston, NM, to include the different eleva- tions encountered. MAT ranges from 11.3°C to 15.6°C, Annual Range 19.5°C to 22°C, Annual Precipitation 24-32 cm, seasonally winter dry and with terrain elevations from 1,400-2,000 m. In general, IIC is drier and at a greater elevation than HE. The Eastern and Central sections, although different geographically, are fairly similar with regards to climate, and quite similar when the climates are adjusted for differences in elevation. However, only three taxa are in common, the widespread and tolerant G. minuscula and the two Holospira. The Western section, IIW, is composed of five taxa that have been displaced almost exactly south to a center of sympatry in southern California and Baja California, a dis- placement of some 16° to 17° of latitude (—1,800 km). This section includes the isolated western band of the genus Radiocentrum, the eastern band of which occurs in section IIC, the genus Ashmunella, also found Group IIC, and the seem- ingly ubiquitous G. pellucida. These taxa seem well adapted to a summer dry climate, Mediterranean, with MAT 14°-18°C, Annual Range 10°-14°C, Annual Precipitation 36-56 cm, and at moderate elevations, generally less than 1,000 m. Group III. Humid Subtropical/Humid Continental, Ca, Da. Group III, the Tem- perate climate group, is composed of 15 terrestrial taxa whose modem analogs 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 89 Table 6. — Kishenehn l.f, terrestrial environmental groups. () Western distribution only. Cabbage Patch fauna taxa; W = Humid Tropi- Dry Tropical/Sub Trop Temperate cal Aw, As BSh-BWh BSh-Cs Caf-Daf BS Group II Group III terrestrial taxa Group I e c w Humid Dry CERESIDAE Tozerpina butt si parva n.ss. X T. lends n.s. X HELICINIDAE Waldemaria monticola Lucidella salishora n.s. X (X) L. Columbiana n.s. X SUCCINEIDAE Oxyloma? kintlana n.s. Omalonyx cocleare n.s. X X PUPILLIDAE Pupoides costatus n.s. X P. tephroides X Gastrocopta miniscula G. kintlana n.s. X X (X) X X G. akokala n.s. G. leonardi (X) X Vertigo consteniusi n.s. X X V. cf. V. arenula X (X) VALLONIIDAE Vallonia kootenayorum n.s. X w HAPLOTREMATIDAE Haplotrema simplex X UROCOPTIDAE Holospira tabrumi n.s. X X H. beardi n.s. X X Coelostemma dawsonae n.s. X DISCIDAE Discus mackenziei X Anguispira, cf. A. alternata X LIMACIDAE Deroceras mahiz D. securis ZONITIDAE Nesovitrea pulchra POLYGYRIDAE Praticolella lucifera n.s. Ashmunella sp. n.s. OREOHELICIDAE Oreohelix dawsonae n.s. Radiocentrum kintlana n.s. AMMONITELLIDAE Polygyroidea montivaga Polygyrella, cf. P. polygyrella HUMBOLDTIANIDAE Skinnerelix rothi n.s. HELMINTHOGLYPTIDAE Xerionata constenii n.s. X X (X) (X) (X) w (W) w 90 Annals of Carnegie Museum VOL. 70 are yet living in the United States and southern Canada. All but four of the analogs of the Group III taxa have been displaced eastward, and are typical of the Eastern Division mollusks (sensu Henderson, 1931:182). For these, a center of sympatry would approximate the State of Kentucky, with a climate ranging from Humid, Continental, to Humid, Subtropical, with warm summer. Climatic data for Lex- ington, Kentucky, with possible ranges to include the entire best area of sympatry, are as follows: MAT 12.7°±5°C, Range 25°-30°C, Annual Precipitation —100 cm, little or no dry season, and at moderate elevations of less than 1,000 m. Two of the four terrestrial analogs that are western in modern distribution, occupy a cli- matic region quite similar to the eastern taxa. Polygyroidea is now found at the edge of the Group HW area in its tiny refugium at moderate altitude in the central Sierra Nevada, where the climate is about 6°C warmer than Montana’s Flathead Valley, but with about twice the rainfall. Polygyrella is now found only on isolated mountainous areas having greater than normal rainfall (rainfall windows of Roth and Emberton, 1994), in eastern Oregon, Washington, Idaho, and western Mon- tana. The genus Oreohelix and the Vallonia cyclophorella Group are Western Division {sensu Henderson, 1931:182) taxa that can tolerate somewhat drier en- vironments than the remainder of this group. The Vallonia cyclophorella Group has isolated distribution in the East. Six of the terrestrial taxa of Group III, two of Group IIC, and one aquatic taxon are members of the late Oligocene-early Miocene (Arikareean) Cabbage Patch fauna of southwestern Montana (Pierce and Rasmussen, 1992; Pierce, 1992; 1993). The proposed climatic preference of the Cabbage Patch fauna was: MAT >10°C, January mean temperature >5°C, July mean temperature 15°-20°C, pre- cipitation >50 cm, and elevation >800 meters (Pierce, 1993:990). This is quite computable with the modern climate suggested above for Group III. The aquatic taxa of the Kishenehn fauna (Table 7) have a more western flavor (8 of 19 taxa), and the majority suggest cool water and a temperate climate. Cincinnatia is considered western, with its nearest modern verifiable occurrence nearly 700 km south, and because its sister species, Fonticella, is exclusively western. For aquatic taxa, east/west dispersion, across existing drainage divides, is difficult. North/south dispersion, generally along drainages and flyways, is more common, e.g., the type locality of Valvata humeralis, a common, low altitude northern taxon, is in cool high altitude lakes of central Mexico {fide Bequaert and Miller, 1973:213). Due to this shielding effect of water, climatic data derived from cool freshwater taxa can be misleading. Data derived from taxa requiring warmer waters is more reliable, and informative. Biomphalaria glabrata (subgenus Aus~ tralorbis), the modem analog of B. kishenehnsis, is incapable of reproduction with water temperatures below 20°C, and water temperatures as low as 5°C result in death within a week (Taylor, 1985:269). Biomphalaria havanensis (subgenus Tropicorbis), analog for B. hay deni and B. spira, is presently not found north of 30° North Latitude, which is approximately coincident with the January Mean Temperature isotherm of 10°C. Environmental data derived from freshwater taxa can be useful. From the en- vironmental data suggested by the modem analogs of these taxa, a picture emerges of highlands surrounding or paralleling a valley containing a chain of lakes, some large, some small, some sandy, some mud bottomed, interconnected by a slow- flowing river and or streams, and subject to seasonal flooding and transient ponds, with permanent and semi-permanent water, generally well vegetated to locally almost overwhelmed by vegetation and marshy, but retaining some rocky littoral 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 91 Table 7. — Kishenehn l.f, aquatic environmental groups. W = Western distribution. AQUATIC TAXA Tropical Jan. Mean > 20 C Subtropical Jan. Mean > 10 C Temperate HYDROBIIDAE Cincinnatial howmanana n.s. w Fluminicolal calderense n.s. w VALVATIDAE Valvata procera w V. procera spatiosa n.ss. w PLANORBIDAE Helisoma triangulata n.s. X Planorbella fordensis n.s. X Planorbula, c.f. P. campestris w Biomphalaria haydeni X X B. spira n.s. X X B. kishenehnsis X Menetus textus. n.s. w M. (M.) hilii n.s. w Gyraulus, c.f. G. scabiosus X LYMNAEIDAE Lymnaea bowmanana n.s. X L. lacerta n.s. w L. newmarchi X PHYSIDAE Aplexa, c.f. A. hypnorum X BIVALVES Sphaerium progrediens X Sphaerium (M.) discus n.s. X areas. An earlier warm water episode is suggested by the Biomphalaria spp. Biomphalaria kishenehnsis, clearly a member of the subgenus Australorbis, is a rare component requiring tropical conditions and is, apparently, found only low in the section. Seasonal fluctuation of water levels is suggested by the Biom- phalaria spp., Lymnaea lacerta, and Aplexa hypnorum, all of which are quite tolerant of desiccation. Extralimitality-Terrestrial Taxa. — The molluscan fauna of the early-middle Tertiary of western North America, especially in western Montana where four diverse and significant faunas have now been studied, are notable for an unusual extralimitality of the faunas (Table 8). Genera, subfamilies, even families that are now distributed in widely divergent parts of North America, the Caribbean region. South America, and even Asia are found here in sympatric association during the Tertiary. The cause of this current extralimitality is in major changes in climate, cooling and drying, which occurred during and after the Eocene in the Western Interior of North America, most notably, the Middle Eocene Event (Berggeren and Prothero, 1992; Retallack, 1992; Wolfe, 1992). Since the climate has changed, the modem fauna has changed to adapt to the new climate. Taxa displaced longitudinally, i.e., predominantly EastAVest, are, usually, so displaced by a change in moisture regimes that no longer meet the requirements of the taxa. As a general rule in North America, isohyets [lines depicting equal Mean Annual Precipitation (MAP)] increase eastward from the Rocky Mountains. 92 Annals of Carnegie Museum VOL. 70 ON (N 00 ON o^ ON ’— ' ON 4^ .. ON c 0 d aj NO (N M CO 3' m ON (fl M ON c On ON p 3 ON 0 s S '- t; . ^ .- 00 . ^ 3 a (N "S in cd QC £ ON w., *-> ON w -H (U 0 H ^ ^ ^ ^ # CNO ir^’— I (Noo (N^f^OO mir> OO O m ^ 2 I ^ (U , White River l.f.) = 94.7%; Cd( Kishenehn If <> Three Forks l.f.) = 67.0%; Cd(Three Forks l.f <> White River l.f.) = 62.5%. terrestrial taxa Taxa in common Kishenehn l.f. Three Forks l.f. White River l.f. PUPILLIDAE Pupoides (Ischnopupoides) 2 3 1 Gastrocopta (Albinula) 2 2 G. other s.sp. 2 1 X VALLONIIDAE Vallonia (Cyclophorella Gp) 1 X UROCOPTIDAE Holospira (Haplocion) 2 X POLYGYRIDAE Ashmunella sp. 1 1 OREOHELICIDAE Radiocentrum sp. 1 1 1 AMMONITELLIDAE Polygyrella, cf. polygyrella la 1 1 HUMBOLDTIANIDAE Skinnerelix sp. 1 1 HELMINTHOGLYPTIDAE Xerionata sp. 1 1 ® Modem analog from Group III. temperatures (Group II) for B. haydeni/spira. Both modern Biomphalaria are quite tolerant of seasonal desiccation. Correlation of faunas. — As was previously noted, the mid-Tertiary molluscan faunas of the Western Interior of North America are becoming fairly well known and understood. Within the last 50 years, six molluscan local faunas (l.f.) have been described (Table 8), beginning with Russell’s (1952; 1954) first studies of the Kishenehn Basin. Faunal correlations among these l.f. should result in some insight as to their relative ages. The Kishenehn l.f. is unique in that it is composed of molluscan groups representing the three distinct climatic environments that, apparently, for a short period of time, coexisted temporally in separate, but nearby, geographic environments. For correlation, each Kishenehn l.f. molluscan climatic group was compared with the remaining Western Interior l.f. It was immediately apparent that the Group I mollusks, representing a wet tropical environment, were unique. Except for the broadly tolerant Gastrocopta minus cula, the remaining taxa are unknown among the other l.f. Group II taxa, indicative of a tropical to sub- tropical semiarid environment, correlate well with both the Three Forks l.f. and the White River l.f. Correlations are based upon a modified Coefficient of Dice (Cd) (Sneath and Sokal, 1973:131). It is modified to measure similarities among the three or more faunas used in each study, and is based not on the total taxa of each fauna, but upon the taxa of the three or more faunas that are shared by two or more of the faunas within the designated group. The Three Forks l.f. and White River l.f. fell within the group based on The Kishenehn l.f. group II mollusks (Table 9). Correlation indices from this triplet are as follows: 96 Annals of Carnegie Museum VOL. 70 Table 10. — Correlation of Kishenehn Group HI fauna. Number of species in if as indicated. Cd(Kishenehn If <> Cabbage Patch if) = 84.6%; Cd( Kishenehn l.f <> Deep River l.f.) = 69.6%; Cd( Cabbage Patch l.f <> Deep River l.f.) = 50.0%. terrestrial taxa Kishenehn Taxa in common l.f. Cabbage Patch l.f. Deep River l.f. HELICINIDAE Waldemaria monticola 1 E Hendersonia sp. E 1 PUPILLIDAE Pupoides (Ischnopupoides) 2 1 1 Gastrocopta {s. sir.) 1 1 G. (Albinula) 2 3^’ 1 G. (Vertigopsis) 1 3 VALLONIIDAE Vallonia {Cyclophorella Gp.) 1 1 DISCIDAE Discus sp. 1 1 Anguispira sp. 1 1 LIMACIDAE Deroceras inahiz 1 1 Deroceras securis 1 1 ZONITIDAE Nesovitrea pulchra 1 1 OREOHELICIDAE Oreohelix sp. 1 1 1 Radiocentrum sp. 1 1 AMMONITELLIDAE Polygyroidea montivaga 1 1 ^ From Flint Creek l.f. (Pierce & Rasmussen, 1989). As subgenus Ameralbinula. Kishenehn Gp. II l.f. to White River l.f. — 94.7% Kishenehn Gp. II l.f. to Three Forks l.f. — 67.0% Three Forks l.f. to White River l.f. — 62.5% A second comparison was made using the taxa of Kishenehn Gp. Ill l.f., in- cluding several taxa that could tolerate both Gp. II and Gp. Ill climatic conditions. It was found that the Cabbage Patch l.f. and the Deep River l.f. (Table 10) had significant correlation indices with the Kishenehn l.f.: Kishenehn Gp. Ill l.f. to Cabbage Patch l.f. — 84.6% Kishenehn Gp. Ill l.f. to Deep River l.f. — 69.6% Cabbage Patch l.f. to Deep River l.f.— 50.0% Based on these correlations with and among the relatively well-dated molluscan l.f. of the Western Interior, it can be determined that the Kishenehn Gp. II l.f. could represent ages as old as Uintan or as young as Chadronian, with a Du- chesnean-Chadronian age most probable. Kishenehn Gp. Ill l.f. could range from Orellan to early Barstovian with an Orellan-Arikareean age most probable. The Kishenehn Gp. I l.f. remains enigmatic, in that it bears no resemblance to any other mid-Tertiary fauna of the Western Interior. The tropical, moist climate nec- essary for these mollusks must have predated the semiarid climate of the late 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 97 Duchesnean-Chadronian Ages, and would, most probably, represent the Uintan to early Duchesnean. Conclusions. — The molluscan fauna of the North Fork Kishenehn Basin pre- sents several most interesting problems. This fauna can be clearly separated into three distinct climatic groups, e.g., L Tropical, Wet; IL Tropical-Subtropical, Semi- arid; and III. Temperate, Moist. However, taxa representing Tropical and Tem- perate groups can be found closely associated in the same level at several of the localities. On the face of it, this appears paradoxical, but on examination, three scenarios need to be examined: Scenario 1. The taxa of all three climatic groups did, in fact, coexist in the same area at the same time. However, in examining the taxa involved, and their modern analogs, it becomes apparent that the climates required for many of the taxa are mutually exclusive. For example, it would be most unlikely for a wet tropical-adapted Lucidella or Omalonyx to survive in the moist temperate climate occupied by the Ammonitellidae. Accordingly, this scenario is rejected. Scenario 2. The three climatic groups are separated by time, and represent a history of climatic change for this area and time. As such, an upward progression on outcrop from Group I (low) to Group III (high) would be expected. This is not what was observed. For example, in the lowest units of Locality CMC-1, a robust assemblage of Helicinidae, Ceresidae, and Omalonyx (all Group I) are found associated with Polygrella c.f., P. polygyrella, Nesovitrea pulchra, and two species of albinulid Gastrocopta (all Group III). However, at most localities, there is a general tendency for Group I taxa to be concentrated in lower strata, disap- pearing upwards, and for the diversity and abundance of specimens of Group III to increase upwards, becoming dominant to exclusive near the top at several localities. The evidence observed at any given locality suggests that sedimentation was rapid, probably involving seasonal depositional episodes, and without significant breaks, either of nondeposition or erosion. In fact, unconformities, mature paleo- sols, and faults are not found at any outcrop. Thus, the time interval represented by any given outcrop is, geologically, brief. This agrees well with the findings of Tabrum et al. (1996:307) that, in southwest Montana, most of these intermontane sequences are of limited time span, and, although thousands of feet thick, most are shorter than a million years in duration. In addition, the North Fork of the Flathead River is a strike river. It flows longitudinally through the basin, staying near the axis of the basin, and erodes only the sediments from near the top of the Lower Member of the Kishenehn Formation. Since nearly all localities are a result of erosion by the river, all those localities designated by a NFF symbol (see Figs. 2, 4) are, therefore, of approximately the same age. This suggests that only a limited time interval is recorded by all these localities (Fig. 4). Localities along the tributaries of the North Fork, which tend to enter the river at near right angles, are time transgressive. Localities on tributaries from east of the river are younger, while those from west of the river are older. Unfortunately, these localities are few, and only on the east side of the river. Those along Bowman Creek and Kintla Creek are, indeed, younger based on mollusks and/or vertebrates. The Kishenehn Creek locality, KSH-1, based on cf. Oligoryctes sp. only, could be as young as Orellan. There were no localities on tributaries entering from the west reported herein. However, review of taxa reported by Russell (1952; 1956) from along Couldrey Creek (his C-1 to C-3) in British Columbia, strongly suggest that this 98 Annals of Carnegie Museum VOL. 70 fauna is from a warmer climate and older age (since it includes Biomphalaria kishenehnsis. Group I). The climatic cycles of the Eocene/Oligocene are fairly well known (cf. Proth= ero, 1994; Wolfe, 1992). There were intervals during the Eocene during which climatic change, usually cooling, were rapid and dramatic. A notable cooling event occurred in the middle Eocene (probably middle Duchesnean) (Prothero and Emry, 1996); a second occurred just past the Eocene/Oligocene border, the “Terminal Eocene Event” of Wolfe (1978). It may be possible that a mature, diverse temperate fauna could totally replace a tropical fauna during so short a time interval, but not likely. However, the paradox of tropical and temperate species coexisting in the lower strata of these localities is not resolved. Scenario 3. The three climatic groups represent lowland and upland taxa ex- isting at about the same time but separated by only moderate distance. Early contractile movement along the Lewis Thrust resulted in bold mountains, the ancestral Clark, Lewis, and Livingston Ranges, east of the present Kishenehn Basin area. In these mountains, a mature, diverse temperate gastropod fauna could have had time to develop. Later, as extensional movement began along the Flat- head and other faults, asymmetric basins developed on the west flank of the ancestral Clark Range. Oxygen isotope studies (D. Dettmann, written communi- cation, 1997) have suggested a meltwater contribution to the Eocene Kishenehn lakes; the runoff from a seasonally snow-capped ancestral Livingston Range would be a most plausible source. In the lowlands to the west of the mountains, a Tropical fauna characteristic of the early Eocene Western Interior could linger on as a relict fauna in these well- watered refug ii, despite some cooling and de- creasing precipitation, relying on the runoff waters for survival, while the Sub- tropical, Semiarid dry woodland (Retallack, 1992:382) fauna developed on ad- jacent, drier slopes. In the case of the Kishenehn Basin, paleocurrent studies indicate the ancestral Clark and Livingston Ranges were the primary source of sediment and runoff, resulting in a large, seasonally variable lake or chain of lakes, capable of sustaining wet-tropical vegetation, but surrounded by a semiarid, dry woodland terrain. The Temperate fauna (Group III) was, initially, transported into the basin from the eastern mountains. The empty shells of many terrestrial gastropods, such as the pupillids and valloniids, are known to be superb floaters, difficult to sink even in turbulent water. Larger taxa, such as the ammonitellids, may have been washed in with some body parts intact, resulting in more or less neutral buoyancy. In any event, these shells could arrive with little evidence of abrasion, and would tend to be deposited in windrows along the shorelines of the seasonally fluctuating lake(s). The end climate after the “Terminal Eocene Event” could no longer support the Tropical and Subtropical faunas in the lowlands, but they could have been replaced rapidly by the Temperate fauna that already existed in the nearby highlands. The assumptions from Scenario 2 that deposition was rapid and nearly contin- uous, that the strata encountered vary in age from oldest near the SW margin of the basin to youngest near the NE margin, and the prejudicial effect of a strike river tending to expose sediments of similar age, are all valid constituents of Scenario 3. Considering the paleotopography known for this area, this scenario is plausible. By early Eocene, at the termination of the contractile stage of tec- tonics, regional topography resembled a regionally-re versed version of the modern northern Amazon Basin/Andes Mountains. The difference in latitudinal wind flow, from the northeast to southwest pattern characteristic of near equatorial north 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 99 latitudes, to the northwest to southeast flow consistent with latitudes of 40°-50° north latitude, would result in ample precipitation to maintain a Tropical wet environment (Amazon Basin analog) west of the well- watered ancestral Conti- nental Divide mountains (Andes analog). The dramatic climatic changes caused by the global cooling episode at the end of the middle Eocene, the Middle Eocene Event (middle Duchesnean, ca. 37-38 Ma) (Prothero and Emry, 1996) severely curtailed tropical and warm adapted taxa of the Western Interior. Possible uplift farther to the west, concurrent with the early extensional phase of tectonics (mid- Eocene), would lead to decreased precipitation in the wet tropical terrain west of the mountains, resulting in the semiarid dry woodland environment of the late Eocene. Further severe cooling during the “Terminal Eocene Event” led to the extinction of the Group I and II faunas at the lower elevations of the Kishenehn Basin, and replacement by the Temperate Group III fauna from the eastern high- lands. Even after this relatively detailed study of the northern Kishenehn Basin, an exact age cannot be unambiguously determined. The single radiometric date of 33 ±1 Ma only establishes a minimum age for the Commerce Creek area, and coincides, approximately, with the “Terminal Eocene Event.” Younger sediments should occur as the footwall boundary is approached. Fossil mammals in asso- ciation at molluscan localities are limited. Those from nearby mammal localities are few but do provide approximate dates. The North American Land Mammal “Ages” shown on Figure 2 at several localities were provided by M. Dawson (personal communication, 1999) and are, admittedly, preliminary. The NFF 12 locality is dated as ?Duchesnean“Chadronian based upon cf. Procaprolagus. The KSH 1 mammal locality age, labeled as ?Chadronian-Orellan, is based on a spec- imen of cf. Oligoryctes. The youngest date, early Arikareean (late Oligocene) for KTL 3 is based on Paciculus sp. These dates do show a definite trend for younger strata on the eastern side of the basin. The probable ages of the Group II and Group III faunas of the Kishenehn Basin (see Correlation of Faunas, above) fit quite well into that scenario. The Group II fauna, which correlates extremely well to the White River l.f., existed in the lowlands west of the mountains during the Chadronian NALMA (post-Middle Eocene Event-pre-Terminal Eocene Event). The Group III taxa, correlating very well to the Arikareean Cabbage Patch l.f., which began to appear near the base of the exposed Kishenehn stratigraphic section during the Chadronian NALMA, is an upland fauna that colonizes and becomes dominant in the lowlands during the late Chadronian-Orellan NALMA’s, replacing Groups I and II faunas, and is the precursor of the typical Arikareean NALMA fauna of the Western Interior, as typified by the Arikareean Cabbage Patch fauna of southwestern Montana. The Group I taxa, which are unique, are probably typical of the Tropical Uintan- Duchesnean (pre-Middle Eocene Event), and had persisted as a relict fauna in isolated suitable environments of the western lowlands into the Chadronian NAL- MA. The complete absence of unionid clams from the strata studied is notable. Russell (1952:125-128; 1955:105) found two unionid species, Elliptio salissiensis Russell, 1952 and Lampsilis dayli Russell, 1952 at only one locality, his W-1 (Russell, op. cit.), which has been subsequently destroyed and/or covered by slumping. From its location, W-1 should be one of the older localities on the North Fork of the Flathead River. Ongoing studies of the older strata of the Middle Fork of the Flathead River show that unionid clams, including Russell’s taxa, are 100 Annals of Carnegie Museum VOL. 70 abundant in those strata. The presence of these unionids demands a through flow- ing river, probably to the east or southeast, draining the Kishenehn Basin during the time of deposition. Absence of unionids in younger strata strongly suggest that either the through flowing river ceased to exist, or, possibly, may have shifted to a westward direction. Strong seasonal variation in water levels during depo- sition of the upper strata could have resulted the local extinction of the unionid clams after the through flowing east-bound river either ceased, or drainage shifted to the west. This should be resolved by the ongoing study of the Middle Fork strata. Summary. — Geology. — The Kishenehn basin is a narrow, normal-fault bounded basin which contains as much as 3,400-4,300 m of middle Eocene-early Miocene (ca. 49-20 Ma) synextensional sedimentary rocks of the Kishenehn Formation. Kishenehn sedimentation was rapid and concurrent with extensional movements on the bounding fault systems; the master fault system which controlled the basin- origin being west-dipping faults of the Flathead fault system. Thus, the Kishenehn Formation provides a record of the history of fault-displacements on the Flathead fault system and the tectonic and biologic evolution of the late Paleogene land- scape of NW Montana and SW British Columbia. Systematic downdropping and rotation of Kishenehn strata along the Flathead listric normal fault imparted an eastward tilt to basinal strata, a tilt that flattens upsection toward the northeast. The result of rapid and near-continuous sedimentation above an active listric nor- mal fault system is a wedge-shaped sedimentary prism that is devoid of intrafor- mation angular-unconformities or well-developed paleosols (i.e., unconformities with long hiatuses in sedimentation). Although the age-record of Kishenehn de- posits spans over 25 m.y., our study of Kishenehn mollusks in the North Fork of the Flathead River sampled only a limited age range, mainly Chadronian. The reason for this is that the river weaves a course only slightly oblique to the axis of the Kishenehn basin. Therefore because of rapid Kishenehn sedimentation rates, only a limited amount of geologic time is represented by the fossil localities examined along the river corridor. The fossil data, however time limited, are critical to reconstruction of ancient Kishenehn landscapes and provides insights into paleotopography, paleoenvironments and paleoecology of a rapidly evolving structural basin that was surrounded by mountain ranges. Summary. — Paleontology. — The molluscan fauna of the northern Kishenehn Basin is exceptionally diverse. The 55 taxa recovered are more than twice as many as the next most diverse molluscan fauna, the Cabbage Patch l.f., of the Western Interior. Of the 55 taxa in this fauna, 32 are new, and described herein, and seven require changes in taxonomy. Four taxa are referred to living species, in that the shells of the fossil specimens are indistinguishable from those of living species. However, this does not necessarily imply that they are biologically iden- tical. Sympatric studies of 52 taxa (three taxa were deemed unusable for that study) of the northern Kishenehn Basin clearly showed the mixed nature of this mollus- can local fauna, overall. Modem analogs of eight taxa (seven terrestrial, one aquatic) clearly require a moist tropical environment for survival, and are current residents of the Caribbean Basin and surrounding continents. These are the Group I fauna. The modern analogs of 18 taxa (16 terrestrial, two aquatic) are presently inhabitants of tropical/semitropical semiarid environments. They are currently found in a broad band from the western Gulf of Mexico across northern Mexico and southern United States to Southern and Baja California. These comprise the 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 101 Group II fauna. Two taxa are found in both Groups I and 11. The remaining analog taxa (15 terrestrial, 16 aquatic) generally fit into a moist, temperate group, Group III. They are most plentifully represented in the eastern United States, for which the general area of Kentucky could be considered the core of their range. Four of the analog terrestrial taxa and eight aquatic analogs included in this Group trend towards a western distribution. However, the climatic requirements of these western terrestrial analogs are sufficiently similar to warrant their inclusion in Group III. Three of the terrestrial analogs of Group III overlap into Group II climatic zones. The degree of extralimitality displayed by mollusks of the northern Kishenehn Basin Lf. follows the pattern suggested by the other molluscan faunas of the Western Interior, i.e., pre- “Terminal Eocene Event” (TEE) faunas have a greater extralimitality (Table 8, 50% or greater), while post-TEE faunas have greater endemicity. In the case of the Kishenehn Basin l.f., all (100%) of the Group I and II analogs are displaced latitudinally (south). The Group I analogs are gen^ erally displaced by 20°-30° of latitude (2,200-3,300 km) to the south, with a strong east component included, whereas the Group II analogs are displaced by 10°-20° of latitude (1,100-2,200 km) to the south with less or no eastward com- ponent. Most Group III analogs are displaced only longitudinally (east-west) into areas of greater MAP. Although the core of their present ranges may be distant, many of these taxa can still be found in northwestern Montana, albeit only in isolated favorable habitats at the limits of their modem range. Typically, the Group III displacement is sufficient to insure 150% to 200% of the current MAP in the Kishenehn Valley (Polebridge, MT, 57.2 cm/yr, see Fig. 2). The molluscan fauna of the northern Kishenehn Basin l.f. is of late Eocene- early Oligocene age, occurring primarily during the Chadronian and Orellan NAL- MA’s. However, the three Groups within this fauna permit inferences to be made covering a wider time span. The Group I tropical fauna is a relic of the tropical Uintan-early Duchesnean NALMAs that preceded the Middle Eocene Event, an episode of strong global cooling that occurred during the middle Duchesnean NALMA. The Group II semiarid fauna developed during the drier, and somewhat cooler, climate of the late Duchesnean-Chadronian NALMAs. The Group III tem- perate fauna was developing in the eastern highlands, bordering the Kishenehn Basin, during the Chadronian NALMA. The early presence of taxa of this Group, in association with Group I and II faunas, was probably a result of transportation in runoff from these eastern highlands (McMechan, 1981; Constenius, 1982). However, as further cooling occurred, leading into, or during the “Terminal Eo- cene Event”, Group III taxa rapidly colonized the lowlands previously occupied by the Group I and II faunas. Acknowledgments This study has benefited materially by interaction with and comments by Barry Roth during prep- aration and in review of the manuscript. B. Carter Hearne, Jr., USGS, ably assisted us the field, and in wide-ranging discussions on geology and tectonics. Mary Dawson, Chris Beard, and Alan Tabrum, CMNH, participated in fieldwork, collecting vertebrate and invertebrate specimens when encountered. A National Geographic Society Grant partially supported this research. Carnegie Museum of Natural History obtained necessary collection permits, and provided some additional reimbursement of ex- penses of the investigators. John Carter, CMNH, personally printed some of the photos for Figures 7- 16. Laura Williams, staff artist for the University of Nebraska State Museum, prepared the recon- structed specimen drawings of Figures 7 and 8. Ray Price and Doug Archibald, Queens University, performed the '^^Ar/^^Ar dating of localities. The U. S. National Museum, Jann Thompson, and the 102 Annals of Carnegie Museum VOL. 70 Geologic Survey of Canada, B. Jean Dougherty, were exceptionally helpful in providing type speci- mens for comparison. We thank William Byrne of Explor Data Ltd. for permission to publish seismic data. Two unknown reviewers, one obviously a molluscan biologist, made very helpful and discerning comments on the manuscript, much of which is incorporated in this final copy. Finally, and very importantly, our thanks to John N. and Leona Constenius, who provided invaluable and unflagging support in all phases of fieldwork for this project. Literature Cited Adams, A. 1861. On some new genera and species of mollusks from the north of China and Japan. Annals and Magazine of Natural History, 3rd Series, 8:299-309. Adams, H., and A. Adams. 1853-1858. The genera of recent Mollusca. J. van Voorst, London, Vol. 1; Vol. 2; Vol. 3. Aguayo, C. G., and M. L. Jaume. 1947-1951. Catalogo de los Moluscos de Cuba, Vol. 1 and 2. Published privately. Ancey, C. F. 1881. De quelques mollusques nouveaux ou peu connus. Le Naturaliste, 1:453. . 1887. The Conchologists Exchange, 1:76. (fide Pilsbry, 1939:63) . 1888. Le Naturaliste, 2nd. sen, 2:188. (fide Pilsbry, 1946:218) Anthony, J. G. 1840. Desriptions of three new species of shells. Boston Journal of Natural History, 3:278-279. Auffenberg, K., and R. W. Portell. 1990. A new fossil land snail (Gastropoda: Pulmonata: Poly- gyridae) from the middle Miocene of northern Florida. Nautilus, 103:143-148. Baird, W. 1863. Descriptions of some new species of shells collected at Vancouver Island and in British Columbia by J. K. Lord, Esq., naturalist to the British North American Boundary Com- mission, in the years 1858-1862. Proceedings of the Zoological Society of London, 1863:66. Baker, F. C. 1909. Description of a new fossil Lymnaea. Nautilus, 23:19-21. . 1911. The Lymnaeidae of North and Middle America, recent and fossil. Chicago Academy of Science Special Publication, 3:1-539. . 1920«. The life of the Pleistocene or glacial period as recorded in the deposits laid down by the great ice sheets. Illinois University Bulletin, 14:1-476. . 1920h. Pleistocene Mollusca from Indiana and Ohio. Journal of Geology, 28:439-457. . 1928. The fresh water Mollusca of Wisconsin, Part 1. Gastropoda. Wisconsin Geological and Natural History Survey Bulletin, 70:1-507. . 1935. Stagnicolo elrodiana, new name for Lymnaea montana Elrod. Nautilus, 49:64. . 1945. The molluscan family Planorbidae. Collation, revision and additions by H. J. Van Cleve. University of Illinois Press, Urbana, Illinois. Baker, H. B. 1922. The Mollusca collected by the University of Michigan-Williamson Expedition to Vera Cruz, Mexico. Parts I, II and III: University of Michigan Museum of Zoology Occasional Paper, 106:1-95. . 1923. The Mollusca collected by the University of Michigan-Williamson Expedition in Ven- ezuela. Part I, Part II: University of Michigan Museum of Zoology Occasional Paper, 137:1-59. . 1926. Anatomical notes on American Helicinidae. Proceedings of the Academy of Natural Sciences of Philadelphia, 78:35-56. . 1928. Minute American Zonitidae. Proceedings of the Academy of Natural Sciences of Philadelphia, 80:1-44. . 1930<3. The land snail genus Haplotrema. Proceedings of the Academy of Natural Sciences of Philadelphia, 82:405-425. . 1930h. New and problematic West American snails. Nautilus, 43:95-100. Bally, A. W, P. L. Goody, and G. A. Stewart. 1966. Structure, seismic data, and orogenic evolution of the southern Canadian Rocky Mountains. Bulletin of Canadian Petroleum Geology, 14:337- 381. Bartsch, P. 1904. Smithsonian Miscellaneous Collection, 47:13. (fide Pilsbry, 1940:932.) . 1906. The urocoptid mollusks from the mainland of America in the collection of the United States National Museum. Proceedings U. S. National Museum, 3 1(1483): 109-160. Beck, H. H. 1837. Index Molluscorum praesentis aevi, Museo Principis augustissimi Christian! Fred- eric! . . . Fasciculcus primus et secundus Mollusca Gastropoda Pulmonatus. Pp. 124(8). 4to. Haf- niae. Benson, W H. 1856. New land shells collected by E. L. Layard, Esq. and described by W. H. Benson, Esq. Annals and Magazine of Natural History, series 2, 18:433-439. Bequaert, j. C., and W. B. Miller. 1973. The Mollusks of the Arid Southwest, with an Arizona check list. University of Arizona Press, Tucson, Arizona. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 103 Berggren, W. a., and D. R. Prothero. 1992. Eocene-Oligocene climatic and biotic evolution: An overview. Pp. 1-28, in Eocene-Oligocene climatic and biotic evolution (D. R. Prothero and W. A. Berggren, eds.). Princeton University Press, Princeton, New Jersey. Binney, a. 1843. Critical notice of the species found in the United States, which, at present, are described as constituting the genus Pupa. Proceedings of the Boston Society of Natural History, 1:104-106. Binney, W. G. 1858. Proceedings of the Academy of Natural Sciences of Philadelphia, 1857:187. (fide Pilsbry, 1939:218.) Bishop, M. J. 1980. Helicinid land snails with apertural barriers. Journal of Molluscan Studies, 46: 241-246. Bland, T. and J. G. Cooper. 1861. Annals of Lyceum of Natural History of New York, 7:365. Brown, A. P. and H. A. Pilsbry. 1914. Fresh-water mollusks of the Oligocene of Antiqua. Proceed- ings of the Academy of Natural Sciences of Philadelphia, 66:209-213. Burch, J. B. 1972. Freshwater Sphaeriacean Clams (Mollusca: Pelecypoda) of North America. Biota of Freshwater Ecosystems Identification Manual, Environmental Protection Agency, 3:1-31. ■ 1989. North American Freshwater Snails. Malacological Publications, Hamburg, Michigan. Carpenter, P. P. (1856)1857. Monograph on the shells collected by T. Nuttall, Esq., on the California coast, in the years 1834-1835. Proceedings of the Zoological Society of London, 24:209-229. Charpentier, N. D. de. 1837. Catalogue des Mollusques terrestres et fluviatiales de la Suisse. Neue Denkschrift, Allgemeine Schweizerische Gesellschaft Naturwiss, 1(2):28. Cheatum, E. P, and R. W. Fullington. 1971. The aquatic and land Mollusca of Texas. Part 1, The Recent and Pleistocene members of the gastropod family Polygyridae of Texas. Dallas Museum of Natural History Bulletin, 1:1-74. Clarke, A. H. 1973. The freshwater mollusks of the Canadian interior basin. Malacologia, 13:1-509. Clessin, S. 1882. Die Familie der Ancylinen. In Abbildungen nach der Natur mit Beschreibungen. Systematisches Conchylien-Cabinet von Martini und Chemnitz, 1(6): 1-80. Bauer und Respe, Niirnberg, Germany. Cockerell, T. D. A. 1914. Tertiary Mollusca from New Mexico and Wyoming. American Museum of Natural History Bulletin, 33:101-107. Cockerell, T. D. A., and J. Henderson. 1912. Mollusca from the Tertiary Strata of the West. American Museum of Natural History Bulletin, 31:229-233. Coney, P. J. 1987. The regional tectonic setting and possible causes of Cenozoic extension in the North American Cordillera. Pp. 177-186 in Continental Extensional Tectonics. (M. P Coward, J. F. Dewey, and P L. Hancock, eds.). Geological Society of America Special Publication 28. Constenius, K. N. 1981. Stratigraphy, sedimentation, and tectonic history of the Kishenehn Basin, northwestern Montana. Unpublished Masters Thesis, University of Wyoming, Laramie, Wyoming. = — — . 1982. Relationship between the Kishenehn Basin and the Flathead listric normal fault system and Lewis thrust salient. Pp. 817-830, in Geologic studies of the Cordilleran Thrust Belt (R. B. Powers, ed.). Rocky Mountain Association of Geologists. -. 1988. Structural configuration of the Kishenehn Basin delineated by geophysical methods, northwestern Montana and south-eastern British Columbia. The Mountain Geologist, 25:13-28. . 1989. Geology and Paleontology of the Kishenehn Basin, northwestern Montana. Pp. 1-21, in Kishenehn Basin Study Project ’89 (K. N. Constenius, ed.). Published privately. . 1996. Late Paleogene extensional collapse of the Cordilleran foreland fold and thrust belt. Geological Society of America Bulletin, 108:20-39. Constenius, K. N., and J. R. Dyni. 1983. Lacustrine oil shales and stratigraphy of part of the Kishenehn Basin, northwestern Montana: Colorado School of Mines Mineral and Energy Re- sources, 26:1-16. Constenius, K. N., M. R. Dawson, H. G. Pierce, R. C. Walter, and M. V. H. Wilson. 1989. Re- connaissance paleontologic study of the Kishenehn Formation, northwestern Montana and south- eastern British Columbia. Pp. 189-203, in Montana Geological Society, 1989 Field Conference Guidebook: Montana Centennial Edition (D. French and R. F. Grabb, eds.). Montana Geological Society. Cooke, C. M. 1921. Notes on Hawaiian Zonitidae and Succineidae. Occasional Papers of the Bernice P. Bishop Museum, 7:261-277. Cooper, J. G. 1863. Proceedings of the California Academy of Sciences, 3:62. . 1870. American Journal of Conchology, 5:196. Crosse, H., and P Fischer. 1872. Diagnosen Molluscorum novorum, Guatemalae et Republicae Mexicanae Incolarum. Journal Conchliologie, 20:146-147. CuRiALE, J. A., S. W. Sperry, and J. T. Senftle. 1988. Regional source rock potential of Oligocene 104 Annals of Carnegie Museum VOL. 70 Kishenehn Formation, northwestern Montana. American Association of Petroleum Geologists Bul- letin, 72:1437-1449. Dahlstrom, C. D. a. 1970. Structural geology in the eastern margin of the Canadian Rocky Moun- tains. Bulletin of Canadian Petroleum Geology, 18:332-406. Dall, W. H. 1895. On a new species of Holospira from Texas. Nautilus, 8:112. . 1896. Synopsis of the subdivisions of Holospira and some related genera. Nautilus, 9:50- 51. -. 1898. A new Polygyra from New Mexico. Nautilus, Vol. 12:75. . 1905. Land and freshwater mollusks of Alaska and adjoining regions. Smithsonian Institution Harriman- Alaska Expedition, 13:1-171. Daly, R. A. 1912. Geology of the North American Cordillera at the Forty-ninth parallel. Geological Survey of Canada Memoir, 38:1-857. Dawson, G. M. 1875. Land and fresh-water Mollusca, collected during the summers of 1873-74, in the vicinity of the forty-ninth parallel-Lake of the Woods to the Rocky Mountains, in Report on the geology and resources of the region in the vicinity of the forty-ninth parallel, from the Lake of the Woods to the Rocky Mountains, with lists of plants and animals collected, and notes on the fossils (North American Boundry Commission, 1872-1876, ed.). B. Westerman & Co., New York, New York. Deshayes, G-P. 1830. Encyclopedie methodique. Histoire des Vers par Bruguiere et Lamarck, com- pletee par Deshayes, Vol. II. Paris, France. Dickinson, W. R. 1991. Tectonic setting of faulted Tertiary strata associated with the Catalina core complex of southern Arizona. Geological Society Special Paper, 264:1-106. Draparnaud, J. PR. 1801. Tableau des Mollusques terrestres et fluviatiles de la France, viii-164. Dunker, W. B. R. H. 1850. Planorhis, Pp. 35-62, in Systematisches Conchylien-Cabinet (E H. W. Martini, and J. H. Chemnitz, eds.). Vol. 1, part 17. Elrod, M. J. 1902. Collecting shells in Montana. Nautilus, 15:103-104. Evanoff, E., and B. Roth. 1992. The fossil land snail Helix leidyi Hall & Meek, 1855, a member of a new genus of Humboldtianidae (Gastropoda: Pulmonata). Veliger, 35:122-132. Evanoff, E., D. R. Prothero, and R. H. Lander. 1992. Eocene-Oligocene climatic change in North America: The White River Formation near Douglas, east-central Wyoming. Pp. 116-130, in Eo- cene-Oligocene Climatic and Biotic Evolution (D. R. Prothero and W. A. Berggren, eds.). Prince- ton University Press, Princeton, New Jersey. Evans, J., and B. E Shumard. 1854. Descriptions of new fossil species from the fresh water Tertiary Formation of Nebraska, collected by the North Pacific Railroad Expedition under Gov. J. J. Ste- vens. Proceedings of the Academy of Natural Sciences of Philadelphia, 7:164-165. Ferussac, a. E. j. P j. F. 1821-22. Tableaux systematiques des Animaux Mollusques . . . Suivis d’un prodrome general pour tous les Mollusques terrestres ou fluviatiles, vivants ou fossiles. 3 Parts. Part II. Tableau de la famile des Limagons, Paris, France. Fields, R. W., D. L. Rasmussen, A. R. Tabrum, and R. Nichols. 1985. Cenozoic rocks of the intermontane basins of western Montana and eastern Idaho. Pp. 9-36, in Cenozoic paleogeography of west-central United States: Rocky Mountain Section (R. M. Flores and S. S. Kaplan, eds.). Society of Economic Paleontologists and Mineralogists. Finch, V. R., G. T. Trewartha, A. H. Robinson, and E. H. Hammond. 1957. Section A2 Climate Types and their Distribution. Pp. 128-206, in Elements of Geography, Fourth Edition. McGraw- Hill, New York, New York. Fitzinger, L. j. F. j. 1833. Systematisches Verzeichniss der im Erzherzogthune Oesterreich Vorkom- memden Weichthiere, etc. Beitrage zur Landeskunde Oesterreich’s unter der Enns, 3:88-122. Fleming, J. 1820. Mollusca. Pp. 598-635, in Edinburg Encyclopaedia: Edinburg, Vol. 14. Franzen, D. S., and a. B. Leonard. 1943. The Mollusca of the Wakarusa River valley. The Uni- versity of Kansas Science Bulletin, 29:363-439. Gabb, W. M. 1866. Descriptions of three new species of land shells from Arizona. American Journal of Conchology, 2:330-331. Galbreath, E. C. 1969. Gastropods from the Tertiary deposits of northeastern Colorado. Transactions of the Illinois Academy of Science, 62:91-93. Gardner, J. 1945. Mollusca of the Tertiary formations of northeastern Mexico. Geological Society of America Memoir, 11:1-332. Gerber, J. 1996. Revision der Gattung Vallonia Risso 1826 (Mollusca: Gastropoda: Valloniidae). Schriften zur Malakozooligie Heft, 8:1-229. Gill, T. 1863. Systematic arrangement of the mollusks of the family Viviparidae, and others, inhab- iting the United States. Proceedings of the Academy of Natural Sciences of Philadelphia, 15:33- 40. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 105 Goodfriend, G. a., and R. M. Mitterer. 1988. Late Quaternary land snails from the North Coast of Jamaica: Local extinctions and climatic change. Palaeogeography, Palaeoclimatology, Palaeoe- cology, 63:293-311. Gould, A. A. 1846. Expeditions Shells: described for the work of the United States exploring ex- pedition, commanded by Ch. Wilkes, during the years 1838-1842. Boston Society of Natural History Proceedings, 2:166. — . 1847. Descriptions of species of Limniadae from collection of the Exploring Expedition:. Proceedings of the Boston Society of Natural History, 2:214-214. — — . 1852. Mollusca and shells. United States Exploring Expedition, during the years 1838, 1839, 1840, 1841, 1842, under the command of Charles Wilkes, U. S. N., 12:1-510. Gould and Lincoln, Boston, Massachusetts. — — — -. 1855. New species of land and fresh-water shells from western (N.) America. Proceedings of the Boston Society of Natural History, 5:127-130. Gray, J. E. 1840. Manual of land- and fresh- water shells of the British Isles (W. Turton, ed.). New Edition. Longman, Rees, Orme, Brown and Greene, London, United Kingdom. — . 1856. On the position of the genus Proserpina in the system, and a description of its dentition. Proceedings of the Zoological Society of London, 29:99-102. Gregg, W. O., and D. W Taylor. 1965. Fonticella (Prosobranchia: Hydrobiidae), a new genus of west American freshwater snails. Malacologia, 3:103-110. Haldeman, S. S. 1842-45 [1840-71]. A monograph of the freshwater univalve Mollusca of the United States (or A monograph of the limniades or freshwater univelve shells of North America), in 9 parts. J. Dobson, Philadelphia, Pennsylvania. 1847. G. Leptoxis. Leptoxe. Rafinesque. Pp. 1-6, in Illustrations conchyliologiques (J. C. Chenu, ed.). A. Franck, Paris, France. . 1867. Description of a new species of Limnaea. American Journal of Conchology, 3:194. Hall, J., and F. B. Meek. 1855. Descriptions of new species of fossils from the Cretaceous Formations of Nebraska, with observations upon BacuUtes ovatus and B. compressus, and the progressive development of the septa in baculites, ammonites, and scaphites. American Academy of Arts and Sciences Memoirs, 5:379-411. Hanna, G. D. 1920. Fossil mollusks from the John Day Basin in Oregon. University of Oregon Publication, 1:1-8. — — — 1922. Fossil freshwater mollusks from Oregon contained in the Condon Museum of the University of Oregon. University of Oregon Publication, 1:1-23. Hanneman, D. L., and C. J. Wideman. 1991. Sequence stratigraphy of Cenozoic continental rocks, southwestern Montana. Geological Society of America Bulletin, 103:1335-1345. Hanneman, D. L., C. J. Wideman, and J. W. Halvorson. 1994. Calcitic paleosols: Their use in subsurface stratigraphy. American Association of Petroleum Geologists Bulletin, 78:1360-1371. Hemphill, H. 1890. New varieties of western land shells. Nautilus, 4:41-43. Henderson, J. 1931. Molluscan provinces in the western United States. University of Colorado Stud- ies, 18:177-186. 1935. Fossil non-marine Mollusca of North America. Geological Society of America Special Paper, 3:1-313. Herrington, H. B. 1962. A revision of the Sphaeriidae of North America (Mollusca: Pelecypoda). University of Michigan Museum of Zoology Miscellaneous Publications, 118:1-74. Hochberg, F. G., Jr., B. Roth, and W. B. Miller. 1987. Rediscovery of Radiocentrum avalonense (Hemphill in Pilsbry, 1905)(Gastrocopta: Pulmonata). Bulletin Southern California Academy of Science, 86:1-12. Hopkins, W. S. J., and A. R. Sweet. 1976. A microflora from a short section of the Paleogene Kishenehn Formation, southeastern British Columbia. Geological Survey of Canada Paper, 76- lB:307-309. Hubendick, B. 1951. Recent Lymnaeidae. Their variation, morphology, taxonomy, nomenclature, and distribution. Kungl Svenska Vetenskapsakademiens Handlingar, Fjorde Serien. Band, 3:1-223. Hubricht, L. 1985. The distributions of the native land mollusks of the eastern United States. Fiel- diana Zoology, new series, 24:1-191. Ihering, H. von. 1892. Morphologic und Systematik des general Apparates von Helix. Zeitschrift fur Wissenschaft Zoologie, 54:386-520. Jones, P. B. 1969. The Tertiary Kishenehn Formation, British Columbia. Bulletin of Canadian Petro- leum Geology, 17:234-246. Kobelt, W. 1876-1881. Illustriertes Conchlyienbuch. Vol. 1 and 2. Bauer und Respe, Niirenburg, Germany. Koppen, W. 1931. Grundriss der Klimakunde. W. De Gruyter, Berlin, Germany. 106 Annals of Carnegie Museum VOL. 70 Lamarck, J. B. P. de. 1799. Prodrome d’une nouvelle Classification des Coquilles, etc: Memoire de Societe Histoire Naturel, Paris, Vol. i, Prairial, An vii (May 1799):63-91. LaRocque, a. 1960. Molluscan faunas of the Flagstaff Formation of central Utah. Geological Society of America Memoir, 78:1-100. . 1966-1970. Pleistocene Mollusca of Ohio. Ohio Division of Geological Survey Bulletin, 62: 1-800. Lea, I. 1837 (1834). Observations on the Naiades; and descriptions of new species of that, and other families. Transactions of the American Philosophical Society, new series, 5:5-119. -. 1838. Descriptions of new freshwater and land shells. Transactions of the American Philo- sophical Society, 6:1-111. . 1839. Transactions of the American Philosophical Society, 6:87. (fide Pilsbry, 1946:221) . 1841. On freshwater and land shells (cont). Proceedings of the American Philosophical Society, 2(17):30-34. — 1858. Descriptions of a new Helix and two new planorbs. Proceedings of the Academy of Natural Sciences of Philadelphia, 10:41. Leonard, A. B. 1946. Three new pupillids from the lower Pleistocene of central and southwestern Kansas. Nautilus, 60:20-24. 1950. A Yarmouthian molluscan fauna in the Midcontinent region of the United States. University of Kansas Paleontological Contributions, 3:1-48. — . 1946. A review of the genus Tryonia Pilsbry (Gastropoda: Amnicolidae). The University of Kansas Science Bulletin, 31:107-114. Leonard, A. B., and D. Franzen. 1944. Mollusca of the Laverne Formation (lower Pliocene) of Beaver County, Oklahoma. The University of Kansas Science Bulletin, 30:15-39. — . 1946. A review of the genus Calipyrgula Pilsbry (Gastrocopta: Amnicolidae). The University of Kansas Science Bulletin 30, part 1(5): 107-1 14. Link, H. F. 1807. Beschreibung der Naturalien-Sammlung der Universitat zu Rostock, 2/3:82-160. Rostock, Germany. Linne, C. a. 1758. Systema Naturae, per regna tria naturae, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. Tenth Edition. Holmiae, Laurentii Salvii. Mackenzie, J. D. 1916. Geology of a portion of the Flathead coal area, British Columbia. Geological Survey of Canada Memoir, 87:53. Martens, E. von. 1860. Die Heliceen nach natiirlicher Verwandtschaft systematish geordnet von Joh. Christ. Albers, 2nd ed. Berlin, Germany. — 1873. Binnenmollusken Venezuela’s. Festschr. zur Feier des hundertjahrigen Bestehens der Ges. naturf. Fr., 1873:157-225. — 1890-1901. Biologia Centrali-Americana. Part 9, Zoology: Terrestrial and Fluviatile Mol- lusca, 44 Pis. 1890:1-40; 1891:41-96; 1892:97-176; 1893:177-248; 1897:249-288; 1898:289- 368; 1899:369-472; 1900:473-616; 1901:617-706 + i-xxviii. London, United Kingdom. McKenna, M. C., P. Robinson, and D. W. Taylor. 1962. Notes on Eocene Mammalia and Mollusca from Tabernacle Butte, Wyoming. American Museum Novitates, 2102:33. McMechan, R. D. 1981. Stratigraphy, sedimentology, structure and tectonic implications of the Ol- igocene Kishenehn Eormation, Elathead Valley graben, southeastern British Columbia. Unpub- lished Ph.D. Thesis, Queen’s University, Kingston, Ontario. McMechan, R. D., and R. A. Price. 1980. Reappraisal of a reported unconformity in the Paleogene (Oligocene) Kishenehn Formation: Implications for Cenozoic tectonics in the Flathead Valley Graben, southeastern British Columbia. Bulletin of Canadian Petroleum Geology, 28:37-45. Meek, F. B. 1860a. Descriptions of new organic remains from the Tertiary, Cretaceous and Jurassic rocks of Nebraska. Proceedings of the Academy of Natural Sciences of Philadelphia, 12:175- 185. . 1860/?. Description of new fossil remains collected in Nebraska and Utah by the exploring expedition under the command of Capt. J. H. Simpson. Tertiary Species. Proceedings of the Academy of Natural Sciences of Philadelphia, 12:312-315. . 1873(1872). Preliminary paleontological report, consisting of lists and descriptions of fossils, with remarks on the ages of the rocks in which they are found . . . Pp. 429-518, in U. S. Geo- logical and Geographical Survey of the Territories (Hayden Survey), Sixth Annual Report. USGPO, Washington, D. C. . 1876. Report on the invertebrate Cretaceous and Tertiary fossils of the upper Missouri country. (Species of fresh- and brackish-water lignite beds. Mollusca. Gastropoda). Pp. 531-605, in Ninth Annual Report of the U. S. Geological and Geographical Survey of the Territories, F. V. Hayden, U. S. Geologist-in-Charge, Vol. IX. USGPO, Washington, D. C. Meek, F. B., and F. V Hayden. 1856. Descriptions of new species of Acephala and Gasteropoda 2001 Pierce and Constenius— Eocene-Oligocene Nonmarine Mollusks 107 from the Tertiary formations of Nebraska Territory, with some general remarks on the geology of the country about the sources of the Missouri River. Proceedings of the Academy of Natural Sciences of Philadelphia, 8:111-126. — 1857. Descriptions of new species and genera of fossils collected by Dr. E V. Hayden in Nebraska Territory, under the direction of Lt. G. K, Warren, of U. S. Topographical Engineers; with some remarks on the Tertiary and Cretaceous Formations of the Northwest, and the paral- lelism of the latter with those of portions of the United States and Territories. Proceedings of the Academy of Natural Sciences of Philadelphia, 9:117-148. Menke, C. J. 1830. Synopsis methodica molluscorum generum omnium et specierum earam, quae in Museo Menkeano adservantur; cum synoeymia critica et novarum specierum diagnosibus. Edito altera. 168 p. Pyrmonti, H. Gelpke. — — — . 1847. Vier neue Arten der Gattung Cylindrella Pfr. Zeitschrift fiir Malakozoologie, 4:1-3, Cassel. Morch, O. a. L. 1864. Synopsis Mulloscorum terrestrium er fluviatilium Danae.-Fortegnelse over de i Danmark Forkommende Land- og Ferskvandsbloddyr. 105 p. Kjdbenkavn. Morelet, P~A. 1849-1851. Testacea novissima insulae Cubanae et Americae Centralis. Pars 1 (1849); Pars 2 (1851). J.-B. Bailliere, Paris, France. Moricand, S. 1833. Memoire Societe Physique et d’Histoire Naturel de Geneve, 6:537. {fide Pilsbry, 1940:694) Morse, E, S. 1864. Observations on the terrestrial Pulmonifera of Maine, including a catalog of all the species of terrestrial and fluviatile Mollusca known to inhabit the State, Portland Society of Natural History Journal, 1:1-63. Muller, O. F. 1774. Vermium terrestrium et fluviatilum, seu aeimalium Infusorium, Heiminthicorum et Testaceoram non marinoram succinta historia. Volumen Altemm. Heineck et Faber, Haviniae et Lipsiae. Newcomb, W. 1864. Proceedings of the California Academy of Sciences, 3:118. {fide Pilsbry, 1946: 204.) Patterson, C. M. 1971. Taxonomic studies of Succineidae. Malacological Review, 4:131-202. Pfeiffer, L. 1839. Bericht iiber die ergenbrisse meieer reise nach Cuba, im winter 1838-1839. Weig- mann's Archiv fur Naturgeschichte, 9:346-358. — — — . 184l£i-1846. Symbolae ad Historiam Heliceorum. Vol. I, 1841; II, 1842; III, 1846. Cassellis. — — . 18411?. Symbolae ad Historiam Heliceorum. Cassellis. . 1849. Descriptions of twelve new species of Vitrina and Succinea, from the collections of H. Cummings, Esq. Proceedings of the Zoological Society of London, 1849:1-133. ■ — 1854, Zur molluskenfauna der Insel Cuba: Malakozoologie Blatter, 1:170-213. Cassellis. Pierce, H. G. 1975. Diversity of late Cenozoic gastropods on the Southern High Plains. Unpublished Ph.D. Dissert., Texas Tech University, Lubbock, Texas. — ^ 1992. The nonmarine mollusks of the late Cenozoic-early Miocene Cabbage Patch fauna of western Montana. IL Terrestrial gastropod families other than Pupillidae (Pulmonata: Stylomma- tophora). Journal of Paleontology, 66:618-628. — — 1993. The nonmarine mollusks of the late Oligocene-early Miocene Cabbage Patch Fauna of western Montana. III. Aquatic mollusks and conclusions. Journal of Paleontology, 67:980-993. . 1996. On sinistral coiling among fossil North American Lymnaeidae. Veliger, 39:220-225. Pierce, H. G., and D. L. Rasmussen. 1989. New land snails (Archaeogastropoda, Helicinidae) from the Miocene (early Barstovian) Flint Creek beds of western Montana. Journal of Paleontology, 63:846-851. . 1992. The nonmarine mollusks of the late Oligocene-early Miocene Cabbage Patch fauna of western Montana. 1. Geologic setting and the family Pupillidae (Pulmonata;Stylommatophora). Journal of Paleontology, 66:39-53. Pilsbry, H. A. 1889. New and little known American mollusks. No.l: Proceedings of the Academy of Natural Sciences of Philadelphia, 41:81-88. — — — . 1890. Note on a southern Pupa. Proceedings of the Academy of Natural Sciences of Phil- adelphia, 42:44. • — — — . 1891. Land and fresh-water mollusks collected in Yucatan and Mexico. Proceedings of the Academy of Natural Sciences of Philadelphia, 43:310-334. . 1895. Manual of Conchology, Index to Helices, Vol. 9:xxxii;124. Academy of Natural Sci- ences of Philadelphia, Philadelphia, Pennsylvania 1902. Manual of Conchology, Second Series: Pulmomata Vol. 15, Academy of Natural Sciences of Philadelphia, Philadelphia, Pennsylvania. — — — . 1903. General Notes. Nautilus, 17:131. 108 Annals of Carnegie Museum VOL. 70 . 1904. Manual of Conchology, Second Series: Pulmonata Vol. 16. Academy of Natural Sci- ences of Philadelphia, Philadelphia, Pennsylvania. . 1905. Mollusca of the Southwestern States, 1: Urocoptidae; Helicidae of Arizona and New Mexico. Proceedings of the Academy of Natural Sciences of Philadelphia, 52:211-290. ■ . 1913. Notes on some Lower Californian helices. Proceedings of the Academy of Natural Sciences of Philadelphia, 65:380-393. . 1916-1918. Manual of Conchology, Second Series: Pulmonata, Vol. 24, Pupillidae. Academy of Natural Sciences of Philadelphia, Philadelphia, Pennsylvania. . 1923. On Triodopsis harfordiana. Nautilus, 37:133-135. . 1926. Manual of Conchology, Second Series: Pulmonata, Vol. 27, Pupillidae. Academy of Natural Sciences of Philadelphia, Philadelphia, Pennsylvania. 1928. Review of the species of Lucidella belonging to the subgenus Poeniella (Helicinidae) of Haiti and Santo Domingo. Proceedings of the Academy of Natural Sciences of Philadelphia, 80:479-482. . 1934a. Mollusks of the fresh- water Pliocene beds of the Kettleman Hills and neighboring oil fields, California. Proceedings of the Academy of Natural Sciences of Philadelphia, 86:541- 570. . 1934Z?. Pliocene fresh- water fossils of the Kettleman Hills and neighboring Californian oil fields. Nautilus, 48:15-17. . 1934(1'. Review of the Planorbidae of Florida, with notes on other members of the family. Proceedings of the Academy of Natural Sciences of Philadelphia, 86:29-66. ■ . 1939-1940. Land Mollusca of North America (North of Mexico). Academy of Natural Sciences of Philadelphia, Monograph 3, Vol. 1(1-2): 1-994. . 1946-1948. Land Mollusca of North America (North of Mexico). Academy of Natural Sciences of Philadelphia, Monograph 3, Vol. 2(1-2): 1-1 1 13. . 1953. Inland Mollusca of northern Mexico. 11. Urocoptidae, Pulillidae, Strobilopsidae, Val- loniidae and Cionellidae. Proceedings of the Academy of Natural Sciences of Philadelphia, 105: 133-167. PiLSBRY, H. A., AND T. D. A. CocKERELL. 1899. A new genus of helices. Nautilus, 12:107. PiLSBRY, H. A., T. D. A. Cockerell, and E. G. Vanatta. 1899. Proceedings of the Academy of Natural Sciences of Philadelphia, 50:227. Preston, H. B. 1910. Additions to the non-marine molluscan fauna of British and German East Africa and Lake Albert Edward. Annals and Magazine of Natural History, Eighth Series, 6:526- 536. Price, R. A. 1965. Elathead map area, British Columbia and Alberta. Geological Survey of Canada Memoire, 336:221. Prime, T. 1860. Synonymy of the Cyclades, a family of acephalous Mollusca. Part 1: Proceedings of the Academy of Natural Sciences of Philadelphia, 1860:267-301. Prothero, D. R. 1994. The Eocene-Oligocene Transition: Paradise Lost. Columbia University Press, Columbia, New York. Prothero, D. R., and R. J. Emry. 1996. Summary. Pp. 664-683, in The terrestrial Eocene-Oligocene transition in North America (D. R. Prothero and R. J. Emry, eds.). Cambridge University Press, Cambridge, New York. Rafinesque, C. S. 1815. Analyse de la Nature, ou tableau de I’univers et des corps organises. Palerme. . 1819. Prodrome de soixante et dix generes nouveaux. Journal de Physique, de Chimie, d’Histoire Naturel, 88:425. . 1820. Annual synopsis of new genera and species of animals, plants, &c. discovered in North America. Annals of Nature, 1:1-25. Rasmussen, D. L. 1973. Extension of the middle Tertiary unconformity into western Montana: North- west Geology, 2:27-35. Retallack, G. j. 1992. Paleosols and changes in climate and vegetation across the Eocene/Oligocene boundary in North America. Pp. 382-398, in Eocene-Oligocene Climatic and Biotic Evolution (D. R. Prothero and W. A. Berggren, eds.). Princeton University Press, Princeton, New Jersey. Richards, C. S. 1963. Apertural lamellae, epiphragms, and aestivation of planorbid mollusks. Amer- ican Journal of Tropical Medicine and Hygene, 12:154-263. Richards, H. G. 1937. Land and freshwater mollusks from the Island of Cozumel, Mexico, and their bearing on the geologic history of the region. Proceedings of the American Philosophical Society, 77:249-262. Risso, A. 1826. Histoire naturelle des principales productions de I’Europe meridionale, et particuli- eremente de cedes des environs de Nice et des Alpes-Maritimes. Vol. 4:439. Paris et Strasburg, France. 2001 Pierce and Constenius — Eocene-Oligocene Nonmarine Mollusks 109 Robinson, G. D. 1960. Middle Tertiary unconformity in southwest Montana: U.S. Geological Survey Professional Paper, 400-B: 227-228. Ross, C. P. 1959. Geology of Glacier National Park and the Flathead region, northwestern Montana. U. S. Geological Survey Professional Paper, 296:1-125. Roth, B. 1984. Lysinoe (Gastropoda: Pulmonata) and other land snails from Eocene-Oligocene of Trans-Pecos, Texas, and their climatic significance. Veliger, 27:200-218. — — ■. 1986. Land mollusks (Gastrocopta:Pulmonata) from early Tertiary Bozeman Group, Mon- tana. Proceedings of the California Academy of Sciences, 44:237-267. . 1991. A phylogenetic analysis and revised classification of the North American Haplotre- matidae (Gastropoda: Pulmonata). American Malacological Bulletin, 8:155-163. . (in press). Upper Cretaceous (Campanian) land snails (Gastropoda: Stylommatophora) from Washington and California. Journal of Moiluscan Studies, 66(3):(). Roth, B., and K. C. Emberton. 1994. “Extralimital” land mollusks (Gastropoda) from the Deep River Formation, Montana: Evidence for mesic medial Tertiary climate. Proceedings of the Acad- emy of Natural Sciences of Philadelphia, 145:93-106. Roth, B., and P. K. M. Megaw. 1989. Early Tertiary land mollusks (Gastropoda: Pulmonata) from Sierra Santa Eulalia, Chihuahua, Mexico, and the origins of the North American arid-land mollusk fauna. Malacological Review, 22:1-16. Russell, L. S. 1926. Mollusca of the Paskapoo Formation in Alberta. Transactions of the Royal Society of Canada, Third Series, 20:207-220. — ■. 1929. Upper Cretaceous and lower Tertiary Gastropoda from Alberta. Transactions of the Royal Society of Canada, Third Series, 23:81-90. ■ — — . 1931. Early Tertiary Mollusca from Wyoming. Bulletins of American Paleontology, 18:3- 39. ■ — — . 1938. New species of Gastropoda from the Oligocene of Colorado. Journal of Paleontology, 12:505-507. ■ — — — . 1952. Moiluscan fauna of the Kishenehe Formation, southeastern British Columbia. National Museum of Canada Bulletin, 126:120-141, 1954. Mammalian fauna of the Kishenehn Formation, southeastern British Columbia. Na- tional Museum of Canada Bulletin,! 32:92-1 1 1. . 1956a. Additions to the moiluscan fauna of the Kishenehn Formation, southeastern British Columbia and adjacent Montana. National Museum of Canada Bulletin, 136:102-119. . 19561?. Non-marine Mollusca from the North Park Formation of Saratoga Valley, Wyoming. Journal of Paleontology, 30:1260-1263. — — — . 1964. Kishenehn Formation. Bulletin of Canadian Petroleum Geology, 12:536-543. Say, T. 1816. Conchology. Nicholson’s British Encyclopedia of Arts and Sciences, 1st (American) Ed., species No.4, pl.l, fig. 2, unpaged. ■ . 1817. Conchology. Nicholson’s British Encyclopedia of Arts and Sciences, 1st (American) Ed., Vol. ii no. 2, unpaged. — — . 1818. Account of two new genera, and several new species, of freshwater and land shells. Journal of the Academy of Natural Sciences of Philadelphia, Vol. 1, Part 2:276-284. . 1821. Descriptions of univalve shells of the United States. Journal of the Academy of Natural Sciences of Philadelphia, 2:149-179. . 1822. Description of univalve terrestrial and fluviatile shells of the United States. Journal of the Academy of Natural Sciences of Philadelphia, 2:370-381. — ■■ 1829. Description of some new terrestrial and fluviatile shells of North America. New Har- mony Disseminator, 2:229-356. ScHLOTHEiM, E. F. 1820. Die Petrefaktenkunde auf ihrem jetzigen Standpunkte durch die Beschreibung seiner Sammlung versteinerter und fossiler Uberreste des Thier- und Pflanzenreichs der Vorwelt erlautert. Beckersche Buchhandlung, Gotha, Germany. ScopoLi, J. A. 1777. Introductio ad historium naturalem sistentes genera lapidum, plantarum et ani- malum, hacentus detecta, caracteribus essentialibus donata, in tribus divisa, subinde ad leges naturae. Prague, Czech Republic. Smith, A., W. B. Miller, C. C. Christensen, and B. Roth. 1990. Land Mollusca of Baja California, Mexico. Proceedings of the California Academy of Sciences, 47:95-158. Sneath, P. H. a., and R. R. Sokal. 1973. Numerical Taxonomy. Freeman, San Francisco, California. SowERBY, G. B. 1822-1834. The genera of recent and fossil shells. Part 4. Stirling, London, United Kingdom. Stearns, R. E. C. 1902. Fossil shells of the John Day region. Science, new series, 15:153-154. Sterki, V. 1889. Nautilus, 3:37, 96, 119. (fide Pilsbry, 1948:883) ■ — . 1890. Notes on northern Pupidae, with descriptions of new species. Nautilus, 3:123—126. 110 Annals of Carnegie Museum VOL. 70 . 1892. Preliminary list of North American Pupidae (North of Mexico). Nautilus, 6:2-8. . 1893. Genus Vallonia Risso. Pp. 247—261, in Manual of Conchology, Second Series: Pul- monata, Vol. 8. Academy of Natural Sciences of Philadelphia, Philadelphia, Pennsylvania. Stimpson, W. 1865. Researches upon the Hydrobiinae and allied forms; chiefly made upon materials in the Museum of the Smithsonian Institution. Smithsonian Magazine, 7:59. Strom, I. A. af. 1767. Beschreibung Norwegischer Insekten, Erstes Sttick. Drontheimischen Gesell- schaft Schriften, 3: 338-395. SwAiNSON, W. 1840. A Treatise on Malacology; or the natural classification of shells and shellfish. London, United Kingdom. Tabrum, a. R., Prothero, D. R., and D. Garcia. 1996. Magnetostratigraphy and biostratigraphy of the Eocene-Oligocene transition, southwestern Montana. Pp. 278-311, in The terrestrial Eocene- Oligocene transition in North America (D. R. Prothero and R. J. Emry, eds.). Cambridge Univer- sity Press, Cambridge, New York. Taylor, D. W. 1954. Nonmarine mollusks from the Barstow Formation of southern California. U. S. Geological Survey Professional Paper, 254-C: 67-80. . 1960. Late Cenozoic molluscan faunas of the High Plains. U. S. Geological Survey Profes- sional Paper, 337:94. . 1966. A remarkable snail fauna from Coahuila, Mexico. Veliger, 9:152-223. . 1975. Early Tertiary mollusks from the Powder River Basin, Wyoming-Montana and adjacent regions. U. S. Geological Survey Open File Report, 75-331:1-515. . 1985. Evolution of freshwater drainages and mollusks in western North America. Pp. 265- 321, in Late Cenozoic history of the Pacific Northwest, interdisciplinary studies of the Clarkia Fossil beds of northern Idaho (C. J. Smiley, A. E. Levitron, and M. Berson, eds.) Allen Press, Lawrence, KS. Thiele, J. 1992. Handbook of Systematic Malacology, Parts 1 and 2. Smithsonian Institution Libraries and The National Science Foundation, Washington. (Translation by J. S. Bhatti of Handbuch der systematischen Weichtierkunde.) Thompson, F. G. 1980. Proseipinoid land snails and their relationships within the Archaeogastropoda. Malacologia, 20:1-33. Tozer, E. T. 1956. Uppermost Cretaceous and Paleocene non-marine molluscan faunas of western Alberta. Geological Survey of Canada Memoir, 280:1-125. Tryon, G. W. 1866. Monograph on the terrestrial Mollusca of the United States. American Journal of Conchology, 2:306-327. Vanatta, E. G. 1911. Mollusca of Arkansas, Louisiana and Mississippi. Proceedings of the Academy of Natural Sciences of Philadelphia, 63:525-532. . 1912. A new species of Vertigo from Florida. Proceedings of the Academy of Natural Sciences of Philadelphia, 64:445. . 1915. Piipoides inornatus new species. Nautilus, 29:95. Van der Velden, A. J., and Cook, F. A. 1994. Displacement of the Lewis thrust sheet in southwestern Canada: New evidence from seismic reflection data. Geology, 22:819-822. Wagner, A. J. 1905. Helicinenstudien. Denkschrift Kaiserliche Akademie Wissenschaft, 77:357-450. Wein. Walker, B. 1915. Apical characters in Somatogvrus with descriptions of three new species. Nautilus, 29:37-41. Westerlund, C. a. 1885. Fauna der in der Palaartischen Region . . . lebenden Binnenconchylien, 1: 88. Lund. White, C. A. 1876. Invertebrate paleontology of the Plateau province, together with notice of a few species from localities beyond its limits in Colorado. Pp. 73-135, in Report on the geology of the eastern portion of the Uinta Mountains (J. W. Powell, ed.) U. S. Geological and Geographical Survey of the Territories, USGPO, Washington, D.C. . 1877a. Bulinus ataviis. Bulletin of the United States Geological and Geographical Survey of the Territories, 3:601. . (1877^)1879. Report on the paleontological field-work for the season of 1877. Pp. 161-272, in Eleventh Annual Report of the United States Geological And Geographical Survey of the Territories enbracing Idaho and Wyoming (F. V. Hayden, ed.). USGPO, Washington, D. C. . (1878)1883. Contributions to invertebrate paleontology. Number 3. Certain Tertiary Mollusca from Colorado, Utah and Wyoming. Pp. 41-48, in U. S. Geological and Geographical Survey of the Territories, Annual Report 12, Part 1, for the year 1878. USGPO, Washington, D.C. . 1886. On the relation of the Laramie molluscan fauna to that of the succeeding fresh-water Eocene and other groups. U. S. Geological Survey Bulletin, 34:1-54. Whiteaves, j. F. 1885. Report on the Invertebrata of the Laramie and Cretaceous rocks of the vicinity 2001 Pierce and Constenius-^ — Eocene-Oligocene Nonmarine Mollusks 111 of the Bow and Belly Rivers and adjacent localities in the Northwest Territory. Canadian Geo- logical Survey, Contribution to Canadian Paleontology, 1:1-89. Wolfe, J. A. 1978. A paleobotanical interpretation of Tertiary climates in the Northern Hemisphere. American Scientist, 66:694-703. 1992. Climatic, floristic, and vegetational changes near the Eocene/Oligocene boundary in North America. Pp 421-436, in The terrestrial Eocene-Oligocene transition in North America (D. R. Prothero and R. J. Emry, eds.). Cambridge University Press, Cambridge, New York. Wolfe, J. A., C. E. Forest, and P. Molnar. 1998. Paleobotanical evidence of Eocene and Oligocene paleoaltitudes in midlatitude western North America. Geological Society of America Bulletin, 110:664-678. Wollaston, T. V, 1878. Testacea Atlantica, or the land and fresh-water shells of the Azores, Madieras, Selvages, Canaries, Cape Verdes and St. Helenas. London, United Kingdom. Yen, T. C. 1944. Notes on fresh-water mollusks of Idaho Formation at Hammett, Idaho. Journal of Paleontology, 18:101-108. . 1946. Eocene nonmarine gastropods from Hot Spring County, Wyoming. Journal of Pale- ontology, 20:495-500. Yoos, T. R., C. J. Potter, J. L. Thigpen, and L. D. Brown. 1991. The Cordilleran foreland thrust belt in northwest Montana and northern Idaho from COCORP and industry seismic reflection data. American Association of Petroleum Geologists Bulletin, 75:1089-1106. Appendix — Molluscan Fossil Localities, Kishenehn Basin, Montana North Fork of the Flathead River, Locality Name Latitude Longitude British Columbia Localities CMC-1 Commerce Creek 49 08' 15" 114 29' 12" Locality CC46 of McMechan, 1981. Age: Late Eocene, ?Duchnesnean to middle Chadronian. North bank of Commerce Creek, 1 .0 km west of Middlepass Creek road. NFF-1 E-5 49 07'46" 114 29'47" Locality E-5 of Russell, 1964. Age: Late Eocene, ?Duchnesnean to middle Chadronian, East bank of Flathead River, 1.0 km north of Sage Creek road bridge. NFF-2 E-4 49 07' 17" 114 29'38" Locality E-4 of Russell, 1964. Age: Late Eocene, Chadronian. East bank of Flathead River, immediately north of Sage Creek road bridge. NFF-3 E-3 49 02'41" 114 29'44" Locality E-3 of Russell, 1964. Age: Late Eocene, middle to late Chadronian. East bank of Flathead River, 1,0 km northeast of Canadian Kootenay Oil Company (d-41-A/82-G22) oil well derrick road. NFF-4 Island 49 02'01" 114 30'00" Age: Late Eocene, late Chadronian. Small island in Flathead River, 3.5 km north of International Boundary. NFF-5 W-2 49 01 '44" il4 30'25" Locality W-2 of Russell, 1964. Age: Late Eocene, middle to late Chadronian. Cut bank 350 m west of Flathead River, 1 .25 km south of mouth of Couldrey Creek. NFF-6 E-2 49 01'18" 114 29'35" Locality E-2 of Russell, 1964, Age: Late Eocene, late? Chadronian. East bank of Flathead River, 2.1 km north of International Boundary. NFF-9 Moose City North 49 00' 36" 114 28'57" Age: Late Eocene, late? Chadronian. West bank of Flathead River, 500 m north of International Boundary. 112 Annals of Carnegie Museum VOL. 70 Montana Localities NFF-10 Moose City South 48 58'34" 114 26'44" Age: Late Eocene, Chadronian. NW/4, 15, 37N, 22 W West bank of Flathead River, 2.25 km south of International Boundary. NFF-12 Starvation Bend 48 55 '39" Age: Late Eocene, Chadronian. C E/2, 36, 37N, 22W West bank of Flathead River, 1.0 km south of mouth of Starvation Creek. 114 23' 14" NFF-17 Kintla Creek/River 48 54'48" 114 22'33" Age: Late Eocene, C N/2, 1, 36N, 22W ?Duchnesnean to Chadronian. East bank of Flathead River, north side of Kintla Creek at mouth. KTL-1 Kintla Creek/CMNH 48 55 '04" 114 26'05" Age: Late Eocene-early Oligocene, NESW/4, 32, 36N, 22W late Chadronian-Orellan. North side of Kintla Creek, at crossing of Inside North Fork Road. NFF-18 Lower Kintla Rapid 48 54'30" 114 22'23" Age: Late Eocene, Chadronian. NENW/4, 12, 36N, 22 W East bank of Flathead River, 1.0 km south of mouth of Kintla Creek. NFF-19 Wurtz Bend 48 54'05" Age: Late Eocene, NW/4, 12, 36N, 22W ?Duchnesnean to Chadronian. East bank of Flathead River, west of Wurtz Landing Strip. 114 22'35" NFF-21 North Ford Creek 48 52 '59" Age: Late Eocene, Chadronian. NW/4, 18, 36N, 21 W East bank of Flathead River, 2.5 km north of mouth of Ford Creek. 114 21 '55" NFF-23 South Ford Creek 48 52 '30" Age: Late Eocene, Chadronian. NW/4, 19, 36N, 21 W East bank of Flathead River, just north of mouth of Ford Creek. 114 21'51" NFF-26 Hook 48 51 '12" Age: Middle to late Eocene, C E/2, 30, 36N, 21 W Duchnesnean to early Chadronian. East bank of Flathead River 800 m north of mouth of Whale Creek. 114 20'52" BWN-2 Bowman Creek 48 47'55" 114 14'25" Age: Late Eocene, Chadronian. SWNE/4, 13, 35N, 21 W North bank of Bowman Creek, 3.0 km from mouth on Flathead River. INSTRUCTIONS FOR AUTHORS ANNALS OF CARNEGIE MUSEUM consists of contributions to the earth sciences (including paleontol- ogy), life sciences, and anthropology, in 30 by 46 pica format (127 by 195 mm or 5 by 7% inches). Submit all manuscripts to the Office of Scientific Publications. Au- thors should give particular attention to scientific con- tent, format, and general style for the ANNALS. Manu- scripts that do not conform to the style of the AN- NALS will be returned to the author immediately. Every manuscript will be subjected to external peer re- view. Authors should submit the names, addresses, phone and fax numbers, and e-mail addresses of at least four qualified potential reviewers for each manuscript. Authors will be asked to subsidize, if funds are available, any or all costs of publication (approximately $ 100/page printed). A detailed set of instructions to authors may be requested from the Office of Scientific Publications. Manuscript Style. — Articles should include the fol- lowing items in this order: title page, abstract, text (with desired headings), acknowledgments, literature cited, ta- bles, figure captions, and copies of illustrations. All manuscripts must be typed double-spaced on standard 814- by 1 1-inch white bond paper, with at least one-inch margins all around, and submitted in quadruplicate — an original for the editors, and three review copies. All pages should be numbered, including tables, literature cited, and the list of figure captions. Only correspon- dence quality or better dot matrix printouts will be ac- cepted; draft mode is unacceptable. A monospaced font, (e.g.. Courier) is required. A disk in WordPerfect x.x or ASCII is helpful. Title Page. The title should be brief, include the ani- mal or plant group involved, be placed on a separate page, and appear two to three inches below the top mar- gin. Include the author’s name(s) and the affiliations of non-CMNH authors. In the case of multiple authorship, indicate the address to which proofs should be sent. Abstracts. Abstracts should be short, but substantive, and included at the head of the first page of text. Text. Do not right justify text or break (hyphenate) a word at the end of a line. Footnotes and acknowledg- ments as footnotes are unacceptable. All text is double- spaced in a monospaced font (e.g.. Courier). Tables and Figure Legends. The list of figure legends and each table should be typed consecutively on indi- vidual pages separate from the text. Tables must be dou- ble-spaced throughout with no vertical lines. Measurements. Metric units should be used, if possi- ble. Literature Cited. All references to literature in text must appear in the Literature Cited section. The data (author, date, and page) in both citations must agree. Do not abbreviate the titles of periodicals or serials. The following style, double-spaced, should be used in Lit- erature Cited: 1) Two authors in a journal series: Soltis, D. E., and P. S. Soltis. 1992. The distri- bution of selfing rates in homosporous ferns. American Journal of Botany, 79:97-100. 2) Same authors repeated — ^use three-em dash: . 1923. The fauna of the Ardyn Obo For- mation. American Museum of Natural History Novitates, 98:1—5. 3) Same authors plus a third author — repeat all authors: Knutson, L. V, R. E. Orth, and W. L. Murphy. 1986. Catalog of Sciomyzidae (Diptera) of America north of Mexico. Entomography, 4: 1-53. 4) Chapter in an edited volume: Rausch, R. L. 1963. A review of the distribution of Holarctic mammals. Pp. 29^3, in Pacific Basin Biogeography (J. L. Gressitt, ed.). Bishop Museum Press, Honolulu, Hawaii. 5) Unpublished dissertation: Smith, J. P. 1976. Review of Eocene Mammals. Unpublished Ph.D. Dissert., University of California, Berkeley, California. 6) Book: White, M. J. D. 1961. The Chromosomes. Meth- euen and Co., Ltd., London, United BCingdom. 7) Journal articles with usual volume and issue number: Anderson, W. I. 1969. Lower Mississippian con- odonts from northern Iowa. Journal of Pale- ontology, 43:916-928. Figures. Including all illustrative materials (line art, halftones, photographs), figures are to be numbered in Arabic numerals. Four sets of figures are required, one (original artwork) for reproduction, three for reviewers. Photocopies of photographs for reviewers are usually not acceptable but are adequate for line drawing review cop- ies. Figures may not be larger than 17 by 12 inches. Reducing figures is the responsibility of the author. All figures must be reducible to a maximum of 127 by 195 mm (30 by 46 picas) without loss of clarity. Line copy should be designed for reduction to % or Vi or actual size. Typewritten figure copy will not be accepted. Pho- tographic figures should be submitted at actual repro- duction size, if possible. Rectangular halftone figures should be abutted, with- out intervening spaces. The printer will insert narrow white spaces during the reproduction process. All figures must have minimally one-inch borders all around. Each figure should be given a protective cover and identified on the back side. Lettering and/or a magnification scale (linear metric scale) for rectangular halftone figures should be placed directly on the photo, not in a blank space between pho- tos. The scale or lettering for closely cropped photos can be placed in blank areas close to the figure. Proof. — The author should answer all queried proof marks and check the entire proof copy. Return corrected page proof with the edited manuscript promptly to the Office of Scientific Publications. If an author chooses to make extensive alterations to a paper in proof stage, the author will bear the cost. Original manuscripts will not be returned \mless request- ed. Figures will be returned to the author only if re- quested prior to publication. ANNALS (^CARNEGIE MUSEUM THE CARNEGIE MUSEUM OF NATURAL HISTORY 4400 FORBES AVENUE • PITTSBURGH, PENNSYLVANIA 15213 VOLUME 70 17 AUGUST 2001 NUMBER 3 CONTENTS ARTICLES A revision of the moth genus Leucania Ochsenheimer in the Antilles (Insecta: Lepidoptera: Noctuidae) Morton S. Adams 179 Middle Eocene Ischyromyidae (Mammalia: Rodentia) from the Shanghuang fissures, southeastern China .... Mary R. Dawson and Banyue Wang 221 REVIEW Trogons and Quetzals of the World (Paul A. Johnsgard) Alejandro Espinosa de los Monteros 231 FROM THE ARCHIVES AND COLLECTIONS Remarks on the Annals of Carnegie Museum's centennial and the introduction of a new feature W. Orr Goehring 233 Editorial from Annals of Carnegie Museum, Volume 1, Number 1 (1901) William J. Holland 235 An excavation in Guanacaste province, Costa Rica David R. Watters and Oscar Fonseca Zamora 237 Editors, ANNALS, BULLETIN and SPECIAL PUBLICATIONS: D. S. Berman, Ph.D. D. R. Watters, Ph.D. J. R. Wible, Ph.D. W. O. Goehring, Managing Editor Manuscripts, subscriptions, orders for individual numbers, and changes of address should be sent to: Office of Scientific Publications Carnegie Museum of Natural History 4400 Forbes Avenue Pittsburgh, PA 15213-4080 Phone (412) 622-3287 Fax (412) 622-8837 E-mail: scipubs@clpgh.org ANNALS OF CARNEGIE MUSEUM is published quarterly by The Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania 15213-4080, by the authority of the Board of Trustees of Carnegie Institute. © 2001 Carnegie Institute. @ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 3, Pp. 179-220 17 August 2001 A REVISION OF THE MOTH GENUS LEUCANIA OCHSENHEIMER IN THE ANTILLES (INSECTA: LEPIDOPTERA: NOCTUIDAE) Morton S. Adams' Research Associate, Section of Invertebrate Zoology Abstract Antillean species of the moth genus Leucania (Lepidoptera, Noctuidae, Hadeninae) are revised. Adult habitus and genitalia of both sexes are described and illustrated for the eighteen species known to occur in the Antilles. Keys are given to adults and genitalia of both sexes. Seven species are described as new: L. toddi and L. educata from Cuba; L. lamisma, L. lobrega, L. sonroja, and L. neiba from Hispaniola; and L. rawlinsi from Jamaica. Lectotypes are designated for L. clarescens Moschler and L. senescens Moschler. Cirphis hampsoni Schaus is placed in synonymy of L. secta Herrich-Schaffer. Leucania dorsalis Walker is removed from synonymy. Key Words: Antilles, Caribbean, Hadeninae, Leucania, Noctuidae, systematics Introduction The systematics of the moth genus Leucania (Noctuidae: Hadeninae) in the Neotropics is in disarray. Adults in this genus are cryptically colored, resembling dried grasses that characterize their habitat. The species are difficult to distinguish based on color or superficial pattern. As a result, species have been poorly char- acterized by authors, are frequently misidentified, and many synonyms have been proposed inadvertently. To resolve these confusing problems, a revision of the New World Leucania was initiated, and this paper constitutes the first of four parts. Revision of the Neotropical species is in preparation, to be followed by review of the Nearctic fauna, and then phylogenetic and biogeographic analysis of the entire New World fauna. Careful delimitation of species resulted from study of morphological details, especially wing patterns and the genitalia of both sexes. Intraspecific geographic variation was determined by examination of material from throughout the range of each species. Historical confusion in nomenclature was resolved by examina- tion of type specimens for all available names. This paper revises the genus Leucania for the Caribbean region (including the Bahamas, Greater Antilles, and Lesser Antilles, but excluding Florida, Bermuda, Tobago, Trinidad, and islands off the northern coast of South America). This treatment provides keys, descriptions, and illustrations for all previously described species, and for seven new species. This revision is a contribution to ongoing entomological research in the Carib- bean region by staff and associates of the Carnegie Museum of Natural History. The alarmingly rapid pace of habitat destruction throughout the Antilles renders urgent the study of the Caribbean fauna. ‘212 High Point Road, West Shokan, New York 12494. Submitted 11 November 1999. 179 180 Annals of Carnegie Museum VOL. 70 History The earliest names for Antillean species were proposed in the 1850s by Guenee and Walker. Guenee (1852) studied material from northeastern South America (present day French Guiana), whereas Walker (1856) examined specimens col- lected in Hispaniola (“St. Domingo” in Tweedie collection, British Museum). These early descriptions are incomplete and often misleading in that they lack illustrations, and do not include diagnostic morphological differences, especially genitalic features. A fundamental work on the Caribbean Lepidoptera fauna, and the source of much subsequent confusion, was that published in 1868 by the influential Euro- pean worker Herrich- Schaffer. His work was based on material from Cuba col- lected by J. Gundlach, and after publication the syntypic material was returned to Gundlach. In a general work on the Lepidoptera of Cuba, Gundlach (1881) re- described the species named by Herrich-Schaffer (1868), but he unfortunately misidentified L. clarescens Moschler, 1890, as L. secta Herrich-Schaffer, 1868. Gundlach’s error was perpetuated by later authors and persisted in the literature until this revision. A series of papers by Moschler (Surinam, 1880; Jamaica, 1886; Puerto Rico, 1890) completed the early descriptive phase for Antillean Leucania. Gundlach was the source of the Puerto Rican material, and in 1891 he redescribed many of the species in Spanish. In that work Gundlach failed to recognize contradictions with his earlier publication in 1881. This further misled subsequent workers who based their interpretation of the species on Gundlach’s publications and not on the original publications of Herrich-Schaffer and Moschler. The modern era of Leucania systematics began with Hampson’s (1905) treat- ment of world species. Hampson’s catalog (1905) was based on the British Mu- seum collection and established usage for many names, including those applied in error that had been accumulating in the literature. Errors in Hampson (1905) were perpetuated at other institutions as the British Museum provided identifi- cation for reference collections developing in the New World. Thus subsequent works all followed Hampson uncritically, including Wolcott (1923, 1936), Draudt (1924), Schaus (1940), Kimball (1965), and Poole (1989). Material Pertinent type material of Herrich-Schaffer in the Gundlach Collection was studied and photographed by J. E. Rawlins in 1990. On the basis of these pho- tographs and associated notes, it has been possible to positively identify the spe- cies described and resolve the confusing historical usage of names. Rawlins re- turned to Cuba in 1992 and borrowed all Leucania material in the collection of the Instituto de Ecologia y Sistematica, Academia de Ciencias de Cuba, including material in the Gundlach Collection, but excluding the types of Herrich-Schaffer from which V. O. Becker (Planaltina, Brazil) had borrowed the abdomens. Becker later dissected the genitalia of these type specimens confirming the interpretation presented in this paper (personal communication, 1995). A series of expeditions to the Dominican Republic by the Carnegie Museum of Natural History (CMNH) provided material from previously unsampled habi- tats at high elevation. These have more than doubled the number of specimens and species available for study from Hispaniola. Other important collections of Leucania from CMNH include material of A. Avinoff and N. Shoumatoff from 2001 Adams — Antillean Moths of the Genus Leucania 181 Jamaica, and recent collections from Dominica, St. Lucia, and Puerto Rico. The author’s collections during the 1980s from Puerto Rico, Jamaica, Guadeloupe, and Barbados have also strengthened this study. This revision benefited from much unstudied material from the Antilles col- lected in the 1970s by staff associated with the United States National Museum of Natural History, Washington, D. C., especially D. R. Davis, O. S. Flint, and the late E. L. Todd. Significant material was obtained from Dominica (1964- 1966), Virgin Islands (1967), Dominican Republic (1973), and St. Lucia, St. Vin- cent, Antigua, and Grenada (1975). This study is based upon examination of 1,577 specimens of Leucania. I thank the following curators of institutional collections and individuals for lending ma- terial: ACC, Institute de Ecologia y Sistematica, Academia de Ciencias de Cuba, Havana. Rafael Alayo. ANSP, Academy of Natural Sciences, Philadelphia. Jon Gelhaus. AMNH, American Museum of Natural History, New York. Frederick H. Rindge. BMNH, Museum of Natural History, London. Alan H. Hayes (deceased). CMNH, Carnegie Museum of Natural History, Pittsburgh, Pennsylvania. John E. Rawlins. CUIC, Department of Entomology, Cornell University, Ithaca, New York. John G. Franclemont. MCZ, Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts. M. Deanne Bowers. ZMHU, Museum fiir Naturkunde an der Humboldt-Universitat, Berlin. H. J. Hannemann. MNHN, Museum National d’Histoire Naturelle, Paris. P Viette. MSA, Private collection of Morton S. Adams, West Shokan, New York. TLM, Private collection of Timothy L. McCabe, Albany, New York. USNM, United States National Museum of Natural History, Washington, D. C. Robert W Poole, Michael G. Pogue. Two specimens (ANSP) have been excluded from this study, both bearing a locality label (“Cuba”) typical of others in the Poey Collection. One was iden- tified as a female of L. Ursula Eorbes, 1936 by dissection. This species is known from southern Canada to North Carolina and west to Illinois in the United States. The second, a male, was identified by dissection as L. lapidaria (Grote), 1875. That species ranges throughout the eastern United States but is not known as far south as peninsular Florida. It is doubtful that either of these species occurs in the Antilles and until their occurrence in Cuba is confirmed by other specimens, they should not be accepted as valid. Measurements are given in millimeters (mm), where there is a range of mea- surements the median is given first followed, in parenthesis, by the minimum and maximum. Terminology of the male genitalia are labeled on Fig. 4 with further details of the valve labeled on Fig. 5C (Forbes, 1954). Terminology of female genitalia are labeled on Figure 13 (Lafontaine, 1987). Key to Adults of Antillean Leucania Based on External Characters 1 . Black dot at end of FW cell (Fig. 2D) 2 1'. No black dot at end of FW cell (Fig. IB). 13 182 Annals of Carnegie Museum VOL. 70 2 (1). FW with all veins white with ground uniform brown, pattern obscure (Cuba) (Fig. 3A) L. toddi 2' . FW veins not entirely white (Fig. 2D) 3 3 (2'). Cubital vein of FW white, at least part of length (Fig. 2D). . 4 3'. Cubital vein of FW not white (Fig. lA) 12 4 (3). Cubital vein of FW white for entire length (Fig. 2D) 5 4'. Cubital vein of FW white for only distal one half (Fig. 2E) 7 5 (4). HW white, FW > 15 mm 6 5'. HW infuscated, FW 13 mm (Cuba) (Fig. 3B) . . . . L. educata 6 (5). FW with longitudinal white line confined to cubital vein, not extending beyond cell (Fig. 2D). . 8 6'. FW with longitudinal white line extending length of cubital vein and beyond end of cell to wing margin (Dominican Republic) (Fig. 3F) L. neiba 1 (4')- FW 15 mm (Hispaniola) (Fig. 2E) . . L. lamisma 7'. FW 13 mm (Cuba and Bahamas) (Fig. 2F) L. secta 8 (3'). FW elongated, apex produced, HW bright white (Fig. 2A) L. chejela 8'. FW not elongated, apex blunt (Fig. 3C) 9 9 (8'). FW ground color pink L. clarescens 9'. FW ground color not pink 10 10 (9'). HW ground color white without infuscation (Fig. 2B) ....................... 11 10'. HW infuscated, especially along veins (Bahamas) (Fig. 3E) L. infatuans 11 (10). FW without dark subcubital shade (Fig. 2B) .............. L. dorsalis 11'. FW with dark subcubital shade (Fig. 2D) . . L. humidicola 12 (10'). Black dot at end of FW cell surrounded by white (Fig. IF) . . L. senescens 12'. Black dot at end of FW cell not surrounded by white (Fig. IE) L. inconspicua 13 (!'). Black shade beyond cell (Fig. 3D) 14 13'. No black shade beyond cell (Fig. ID) 15 14 (13). Black median shade for entire length of FW (Fig. 3D). L. incognita 14'. Black shade only beyond cell (Fig. 2C) L. latiuscula 15 (13'). FW rust color (Hispaniola) L. sonroja 15'. FW ground color brown 16 16 (15'). FW costal cell (basal Va dorsal FW between costa and radius) homogeneous color without definite stripes (Fig. 1C) L. lobrega 16'. FW costal cell with alternating dark and white longitudinal stripes (Fig. IB) ...... . 17 17 (16'). Distribution north of frost-free line along Gulf coast of United States and Bahamas .... L. subpunctata 17'. Distribution south of frost-free line in Antilles, central and northeastern South America L. rawlinsi Key to Antillean Leu can i a Based on Male Genitalia 1. Margin of cucullus concave (Fig. 12 A) L. incognita 1'. Margin of cucullus convex (Fig. 4A) 2 2 (L). Cucullus clavate (Fig. 4A) 3 2'. Cucullus not clavate (Fig. 9A). ........................................ 8 3 (2). Margin of valve produced into sharp point (Fig. 6A) ................ L. inconspicua 3'. Margin of valve rounded (Fig. 7C) 4 4 (3'). Basal sclerite of clasper round (Fig. 1C). L. dorsalis 4'. Basal sclerite of clasper developed into projection (Fig. 4A) 5 5 (4'). Projection from basal sclerite of clasper forming U shape with digitus (Fig. 4A) ..... 6 5'. Projection from basal sclerite of clasper not forming U shape with digitus (Fig. 5A). . . 7 6 (5). Everted aedeagus with appendix on vesica (Fig. 4B) L. subpunctata 6'. Everted aedeagus without appendix on vesica (Fig. 4D) L. rawlinsi 1 (5'). Projection on basal sclerite of clasper angled distally (Fig. 5C). L. sonroja 7'. Projection on basal sclerite of clasper blunt, not angled distally (Fig. 5 A) . ... L. lobrega 8 (2'). Claval area of sacculus modified (Fig. 1 1C) 9 8'. Claval area of sacculus not modified (Fig. 7A) 11 9 (8). Claval area developed into process which extends across face of clasper (Fig. 6C) ..... . .L. chejela 9'. Claval area without elongated process (Fig. 1 1C) ........ 10 2001 Adams — Antillean Moths of the Genus Leucania 183 10 (9'). Claval area developed into sharply serrated projection (Fig. IIC) L. clarescens 10'. Claval area developed into low prominence with sensory hairs (Fig. 9A). . . L. humidicola 11 (8'). Projection of basal sclerite of clasper and digitus form a U shape (Fig. 7A). . L. senescens 11'. Projection of basal sclerite of clasper and digitus not forming a U shape (Fig. 8A). . . 12 12 (11'). Clasper not developed into projection (Fig. 8A) 13 12'. Clasper developed into projection (Fig. 12C) 14 13 (12). Clasper rounded into large plate (Fig. IOC) L. toddi 13'. Clasper with a sharp angle at margin (Fig. 8 A) L. latiuscula 14 (12'). Clasper with blunt projection near digitus (Fig. 12C). . L. infatuans 14'. Clasper with a projection distant from digitus (Fig. 9C) 15 15 (14'). Clasper projection pointed terminally (Fig. 9C) 16 15'. Clasper projection blunt (Fig. 11 A) L. educata 16 (15). Clasper projection straight (Fig. 9C). L. lamisma 16'. Clasper projection angled terminally (Fig. lOA) L. secta 1. 1'. 2 (1'). 2'. 3 (2'). 3'. 4 (3'). 4'. 5 (4). 5'. 6 (4'). 6'. 7 (6). 7'. 8 (7'). 8'. 9 (8'). 9'. 10 (6'). 10'. 11 (10'). 11'. 12 (IT). 12'. 13 (12'). 13'. 14 (13). 14'. 15 (14'). 15'. 16 (13'). 16'. Key to Antillean Leucania Based on Female Genitalia Appendix bursae not developed beyond a simple prominence of corpus bursae (Fig. 13F) L. chejela Appendix bursae developed (Fig. 13A). 2 Appendix bursae arising proximally from ductus bursae (Fig. 14D). ...... L. humidicola Appendix bursae arising distally from end of ductus bursae at junction with corpus bursae (Fig. 14C) 3 Appendix bursae arising from posterior evagination of distal ductus bursae (Fig. 1 3C) . ... L. lobrega Appendix bursae not a posterior pouch from ductus bursae (Fig. 14C) 4 Ductus bursae long and narrow; ratio of length: width greater than 20:1 (Fig. 14A) ... 5 Ductus bursae ratio of length:width less than 20:1 (Fig. 14C) 6 Ductus bursae longer than appendix bursae (Fig. 15C). L. clarescens Ductus bursae and appendix bursae approximately equal in length (Fig. 14A) L. senescens Appendix bursae curving sharply to right (Fig. 14C) 7 Appendix bursae not curving sharply to right (Fig. 13B) 10 Appendix bursae curving cephalad (Fig. 14C) L. latiuscula Appendix bursae curving caudad (Fig. 14B) 8 Appendix bursae thick, equal to width of ductus bursae (Bahamas) (Fig. 15E) L. infatuans Appendix bursae narrow (Fig. 14B) 9 Appendix bursae arising near junction between ductus bursae and corpus bursae (Fig. 14B) L. dorsalis Appendix bursae arising midway along tubular part of bursa copulatrix (ductus bursae in part) (Cuba) (Fig. 15A). . L. toddi Appendix bursae narrow and same length as ductus bursae (Fig. 13 A, 13B) ... L. rawlinsi / L. subpunctata Appendix bursae if narrow, shorter than ductus bursae; if broad then irregularly formed and of variable length (Fig. 13E, 15B) 11 Ductus bursae extends beyond origin of corpus bursae as a narrow tube (Fig. 15B) .... L. educata Ductus bursae an irregular sclerotized stump beyond the corpus bursae (Fig. 13E) ... 12 Appendix bursae same diameter as ductus bursae (Fig. 14F) L. secta Appendix bursae greater diameter than ductus bursae (Fig. 1 3E) ................ 13 Sclerotized portion of appendix bursae one half length of ductus bursae (Fig. 13E) . . 14 Sclerotized portion of appendix bursae less than one quarter length of ductus bursae (Fig. 13D). . 16 Appendix bursae greater than twice the width of ductus bursae (Fig. 13E) L. inconspicua Appendix bursae width less than twice the width of ductus bursae (Fig. 14E) ...... 15 Ostium bursae same diameter as ductus bursae (Fig. 15D) .............. L. incognita Ostium bursae wider than ductus bursae (Fig. 15F) L. neiba Caudad portion of ductus bursae funnel shaped (Fig. 13D). .............. L. sonroja Caudad portion of ductus bursae barrel shaped (Fig. 14E) L. lamisma 184 Annals of Carnegie Museum VOL. 70 Systematic Accounts Genus Leucania Ochsenheimer Leucania Ochsenheimer, 1816:81. Diagnostic Features of Adults. — These moths share the pale tan ground and striate brown shading of many species that rest by day among dead grasses. The usual pattern of the wings is obscure and often reduced to dots. Forbes (1954: 107) gives a discussion of the variation among the species. The genus is limited to a group of moths characterized by distinctive structures of the male genitalia (Franclemont, 1951). A slit divides the unarmed, flap-like cucullus from the re- mainder of the valve, the clasper plate is often modified significantly, the digitus is prominent, a sensory plate is present on the valvula at the base of the cucullus, and sensory hairs are present on the ampulla and editum. The claval area of the sacculus is occasionally modified. A peculiar ‘‘bubble-like” modification of the lower part of the juxta between the bases of the valves is present in this and related genera (Calora, 1966). Coremata are developed in several lineages includ- ing Antillean species. These male secondary sexual characters are large eversible sacs on the ventral aspect of the second abdominal segment. Birch (1972) inter- preted these as distinctive modifications of the brush organs found in many noc- tuid moths, and they are apparently unique to this genus (Varley, 1962). The female genitalia are conservative, and characters have not been found that separate those of Leucania from related genera. Diagnostic Features of Larvae. — The larvae are restricted to feeding on grasses and sedges. In common with other members of this guild (Faronta, Dargida, Aletia and Pseudaletia) they have large, mostly exposed, mandibles and are su- perficially very similar. Shared features of larvae include the following: smooth skin; general color varying from olivaceous green, yellow or pink to brownish; middorsal stripe and body pattern made up of straight longitudinal lines and stripes; head pattern reticulate (Godfrey, 1972:85-99; Bruner et ak, 1975). Sim- ilarity in color may be due to convergence associated with the grass-feeding habit as the closely related Faronta and Dargida differ structurally from Leucania, Aletia, and Pseudaletia (Godfrey, 1972:21). A possible synapomorphy of the lineage containing Leucania, Aletia, and Pseu- daletia was described by Godfrey (1972:11, figs. 208, 212) for a modification of the hypopharynx. Larvae of these genera have the distal hypopharyngeal region covered with long, thin spines which form a brush. This feature is absent in Faronta and Dargida. Crumb (1956:149) was unable to find characters distinguishing Leucania from Aletia and Pseudaletia, but Godfrey (1972:86) separated Leucania from the other two genera on the basis of the position of the frontal punctures. These are on a line connecting bases of the frontal setae in Leucania, but distinctly below that line in Aletia and Pseudaletia. Leucania subpunctata (Harvey) (Fig. lA, 4A-B, 13A) Heliophila subpunctata Harvey, 1875:8. Grote, 1882:30; Grote, 1895:87; Dyar, 1903:162. Leucania subpunctata (Harvey): Smith, 1893:189; Smith, 1903:180. Cirphis latiuscula: Hampson, 1905:542 (in part); Draudt, 1924:167 (in part) (not Herrich-Schaffer, 1868:148) [Misidentifications]. 2001 Adams — Antillean Moths of the Genus Leucania 185 Fig. 1. — Adult Antillean Leucania (LFW = length of forewing): A. L. subpunctata, male, USA, Louisiana, St. Tammany Parish, 6 km NE Abita Springs (MSA, LFW = 18 mm); B. L. rawlinsi, holotype male, Jamaica, St. Andrew Parish, Irishtown (MSA, LFW = 18 mm); C. L. lobrega, holotype male, Dominican Republic, Pedernales, 26 km N Cabo Rojo (CMNH, LFW == 18 mm); D. L. sonroja, holotype male, Dominican Republic, Pedernales, 26 km N Cabo Rojo (CMNH, LFW = 15 mm); E. L. inconspicua, male, Dominican Republic, Puerto Plata, Pico El Murazo, north slope near summit (CMNH, LFW = 15 mm); F. L. senescens, female, Cuba, Pinar del Rio, San Vicente, near mouth of Cueva del Indio, Sierra de los Organos (CMNH, LFW = 15 mm). 186 Annals of Carnegie Museum VOL. 70 Leucania latiiiscula: McDunnough, 1938:77; Franclemont, 1951:63; Franclemont and Todd, 1983:150; Poole, 1989:586 (not Herrich-Schaffer, 1868:148) [Misidentifications]. Diagnosis. — This species is superficially indistinguishable from the widespread L. rawlinsi new species, which replaces it south of the United States border and in the Antilles. The diagnostic character is the presence of an appendix on the vesica of the everted aedeagus. Description. — (Fig. lA). Head: palpi light tan interspersed with black scales; front black with tan scales between antennae. Thorax brown with three contrasting black patagial bands, most posterior broadening into area of white-tipped brown scales; tegulae brown with single row of black dots medially; white-tipped black scales in tuft immediately behind patagia; Forewing: 17 mm (15-18 mm), ground brown with dark shade obscuring cubital vein; small white dot at end of cell; black dot at origin of vein Cu2; postmedial line marked by black dots at veins; usual apical shade accentuated by pale area towards costal margin; fringe brown with contrasting white at veins. Hindwing pearly white with brown scaling on veins and marginal infuscation (greater in females). Underside forewing tan, darker over veins; conspicuous black dot at costal margin at postmedial line. Underside hindwing white with costal and marginal infuscation. Abdomen tan, lighter than wings and thorax; tuft of dark scales on segments one and two; caudal tuft dark brown; coremata not developed. Sexes similar. Male Genitalia (Fig. 4A): uncus, tegumen, and vinculum unmodified; valve with cucullus large, knob-like, otherwise unremarkable. Aedeagus (Fig. 4B) with vesica simple, tubular, bearing appendix about one-third distance from distiphallus, with two long recurved cornuti near primary gonopore. Female Genitalia (Fig. 13A): ductus bursae narrow; appendix bursae similar in shape and size to ductus bursae. Distribution. — This species ranges along the United States coast of the Gulf of Mexico north of the frostfree line. South of this line in Mesoamerica and the Antilles it is replaced by L. rawlinsi. Flight Period. — In the mainland population this species flies throughout the year. The only collection from the Antilles was made in March. Primary Type Data. — Holotype female (BMNH). Verbatim label data: “Helio- phila subpunctata type Harvey” [white label with red border, script] / “495” [small red label] / “Texas, Grote Coll. 81-116” [white label, printed] / “type” [red circular label] / “BM NOCT Slide 7119” [white label, printed]. Material examined. — BAHAMAS. Bimini: no specific locality, 1 d (CMNH). Leucania rawlinsi, new species (Fig. IB, 4C-D, 13B) Cirphis latiiiscula: Hampson, 1905:542 (in part) and pi. 93, fig. 20; Wolcott, 1923:161; Draudt, 1924: 167, pi. 24, row G; Schaus, 1940:186 (not Herrich-Schaffer, 1868:148) [Misidentification]. Diagnosis. — This species is superficially indistinguishable from L. subpunctata Harvey, 1875 which replaces it along the Gulf of Mexico from Texas to Florida. The two species are not known to fly together. The diagnostic character is the absence of an appendix on the vesica of the everted aedeagus in L. rawlinsi, which is present in L. subpunctata (Fig. 4B). It is especially important to distin- guish this species from the broadly sympatric L. lobrega new species, which is also difficult to separate superficially except by microscopic examination of the costal area of the basal one-quarter of the dorsal forewing between the costa and radius. In L. rawlinsi contrasting longitudinal stripes of alternating dark and white are present. These stripes are very obscure or absent in L. lobrega. Differences in the male and female genitalia are detailed in the diagnosis of that species. Description. — (Fig. IB). Head: palpi light tan interspersed with black scales; front black with tan scales between antennae. Thorax brown with three black patagial bands, contrasting, most posterior broadening into area of white-tipped brown scales; tegulae brown with single row of black dots 2001 Adams — Antillean Moths of the Genus Leucania 187 Fig. 2. — Adult Antillean Leucania (LFW = length of forewing): A. L. chejela, female, Dominican Republic, Peravia, 3 km SW Las Nuez, upper Rio Las Cuevas (CMNH, LFW = 17 mm); B. L. dorsalis, male, Dominican Republic, Pedernales, 23.5 km N Cabo Rojo (CMNH, LFW = 15 mm); C. L. latiuscula, male, Dominican Republic, Puerto Plata, Pico El Murazo, north slope near summit (CMNH, LFW =14 mm); D. L. humidicola, female, Dominican Republic, Azua, 8 km NE Padre Las Casas, Rio Las Cuevas (CMNH, LFW = 16 mm); E. L. lamisma, female, Dominican Republic, Pedernales, 5 km NE Los Arroyos (CMNH, LFW = 15 mm); F L. secta, neotype female, Cuba, Allende Mountains (ACC, LFW = 13 mm). medially; white- tipped black scales in tuft immediately behind patagia; Forewing: 17 mm (16-18 mm), ground brown with dark shade obscuring cubital vein; costal area with alternating dark and white longitudinal stripes, central stripe bright white; small white dot at end of cell; black dot at origin of vein Cu2; postmedial line marked by black dots at veins; usual apical shade accentuated by pale area towards costal margin; fringe brown with contrasting white at veins. Hindwing pearly white with 188 Annals of Carnegie Museum VOL. 70 brown scaling on veins and marginal infuscation (darker in females). Underside forewing tan, darker over veins; conspicuous black dot at costal margin at postmedial line. Underside hindwing white with costal and marginal infuscation. Abdomen tan, lighter than wings and thorax; tuft of dark scales on segments one and two; caudal tuft dark brown; coremata not developed. Sexes similar. Male Genitalia (Fig. 4C): uncus, tegumen, and vinculum unmodified; valve with cucullus large, knob-like, otherwise unremarkable. Aedeagus (Fig. 4D) with vesica tubular, simple, unadorned, with- out appendix, terminating in two long recurved cornuti. Female Genitalia (Fig. 13B): as in L. subpunctata; ductus bursae narrow; appendix bursae similar in shape and size to ductus bursae. Distribution. — This widespread species ranges over northeastern South Amer- ica, Central America, and the Antilles. It is replaced along the United States coast of the Gulf of Mexico by L. subpunctata. A Bermudan specimen figured by Ferguson et al. (1991 :fig. 191) was identified as L. subpuncta [sic], a misspelling of L. subpunctata, but appears to be L. rawlinsi based upon the degree of infus- cation of the hind wing. Flight Period. — This species as been collected in January, April, June, July, August and November. Discussion. — The amount of infuscation on the hind wings varies between is- land and continental populations. Specimens from the continent have almost pure white hind wings. The white hindwing form also occurs in Barbados, which may indicate a more recent origin from the mainland of South America. The population on Puerto Rico is very dark. Primary Type Data. — Holotype male (CMNH). Verbatim label data: “JAMAI- CA, St. Andrew, Irishtown, 732 m, 10, 11, 16 Aug. 1988, Morton S. Adams’" [white label, printed] / “HOLOTYPE Leucania rawlinsi Adams” [red paper]. Paratypes.—{A2 6 , 45 9). BAHAMAS. Great Exuma: Simons Point, 23°3rN, 50°75'W, 2 9 (TLM). BARBADOS. St. Michael Parish^ Bridgetown, 1 6 (USNM); St. George Parish: Grove Planta- tion, 2 9 (MSA). CUBA. Matanzas: no specific locality, 1 cJ, 2 9 (USNM). Santiago de Cuba: Sierra Maestra, 305 m, 1 9 (ANSP); no specific locality, 10 3, 2 9 (USNM). DOMINICA. St. Joseph: Grande Savane, 1 9 (USNM). St. Paul: Clarke Hall, 1 9 (USNM). DOMINICAN REPUBLIC. Independencia: 4 km S Los Pinos, Loma de Vientos, 18°35'N, 71°46'W, 455 m, 1 9 (CMNH). La Vega: Constanza, Hotel Nueva Suiza, 1,164 m, 1 S (USNM). Monsenor Nouel: [boundary with La Vega] Loma El Casabito, summit, 19°03'N, 70°31'W, 1,390 m, 2 d, 2 9 (CMNH). Monte Cristi: 5 km NNE Botoncillo, 19°46'N, 71°24'W, 50 m, 1 9 (CMNH). Pedernales: 8 km N Cabo Rojo, 30 m, 1 9 (CMNH); 23.5 km N Cabo Rojo, 18°06'N, 71°38'W, 540 m, 2 d, 1 9 (CMNH); 26 km N Cabo Rojo, 18°06'N, 71°38'W, 730 m, 1 d, 1 9 (CMNH). Puerto Plata: Pico El Murazo, 19°4rN, 70°57'W, 2 d, 1 9 (CMNH). San Juan: 7 km NW Vallejuelo, 18°42'N, 71°16'W, 690 m, 1 d (CMNH). JAMAICA. Portland Parish: 1 mi N Hardware Gap, 1 d (USNM); Port Antonio, 1 d (AMNH). St. Andrew Parish: Chestervale, Yallahs River, 1 d (USNM); Constant Spring, 2 d (CMNH); Irish- town, same label data as holotype, 1 d, 6 9 (CMNH, MSA, USNM; genitalia preparations M. S. Adams, d 2033, d 2041). St. Ann Parish: Moneague, 1 9 (AMNH); Rose Hill, Runaway Bay, 275 m, 1 9 (USNM). Parish unknown: no specific locality, 1 9 (AMNH). PUERTO RICO. Banos de Coamo, 18°03'N, 66°22'W, 5 d, 6 9 (AMNH, CUIC, MSA); Guayama, Aguirre Cent., 1 d (CUIC); Maricao, Hacienda Juanita, 1 9 (MSA); Sabana Grande, 1 d (USNM); San German, 1 9 (CUIC); Toa-Baja, 1 9 (CUIC); Vieques Island, 2 d (CUIC); no specific locality, 1 9 (USNM). ST. CROIX. Mount Eagle, 1 9 (USNM); Kingshill, 4 9 (CUIC, USNM); Orangegrove, West End, 2 d, 1 9 (USNM); 1 mi W airport, 1 d (USNM); no specific locality, 1 9 (USNM). ST. LUCIA. Soufriere: 1 mi NW Soufriere, 1 9 (USNM). ST. MARTIN. Pic du Paradis, 1 d (MSA). ST. VINCENT. Orange Grove, 1 d (USNM). 2001 Adams — Antillean Moths of the Genus Leucania 189 Fig. 3. — Adult Antillean Leucania (LFW = length of forewing): A. L. toddi, holotype male, Cuba, Finca La Ciega, Florida, Camagiiey (USNM, LFW = 13 mm); B. L. educata, male, Cuba, Cuabitas (ACC, LFW = 13 mm); C. L. clarescens, male, Dominican Republic, Monte Cristi, 5 km NNE Botoncillo (CMNH, LFW = 14 mm); D. L. incognita, male, USA, Florida, Dade County, Fuchs Hammock (MSA, LFW = 14 mm); E. L. infatuans, male, USA, Florida, Highlands County, 2 km S Highlands Hammock State Park (CMNH, LFW = 15 mm); E L. neiba, holotype female, Dominican Republic, Elias Pina, North slope Sierra de Neiba 2 km SW Canada, 7 km WSW Hondo Valle (CMNH, LEW = 15 mm). Leucania lobrega, new species (Fig. 1C, 5A-B, 13C) Cirphis microsticha: Schaus, 1940:186 (not Hampson, 1905:529) [Misidentification]. Leucania opalisans: Dickel, 1991 (not Draudt, 1924:164) [Misidentification]. Diagnosis. — This species is most likely to be confused with L. rawlinsi, which 190 Annals of Carnegie Museum VOL. 70 is remarkably similar and sympatric over much of its range. The adult in both sexes can be distinguished by examination under magnification of the costal cell. This is the area at the basal one-quarter of the dorsal forewing between the costal and radial veins. In this species the area has a relatively homogeneous color without definite stripes. In L. rawlinsi, the costal cell has alternating dark and white longitudinal stripes. The central white stripe is especially bright and con- trasting. Although the adults are superficially similar, their genitalia are not. In the male, the clasper is two pronged in L. rawlinsi and blunt without prongs in this species. In the female the appendix bursae opens directly from the distal end of the ductus bursae at the point of origin of the corpus bursae. In L. rawlinsi the appendix bursae arises from the corpus bursae as a long sclerotized tube that expands into a thin-walled chamber before continuing as the ductus seminalis. Description. — (Fig. 1C). Head: palpi gray, flecked with black becoming darker dorsally; front gray- ish tan with dark band between base of antennae. Thorax tan with three dark patagial bands, posterior broadest and blending into tuft of dark scales immediately behind patagia; tegulae grayish tan with fine row of darker hairs medially; disc tan. Forewing: 17 mm (15-18 mm), ground grayish tan, appearing powdery due to fine black scales scattered on ground especially at costa; costal area ho- mogeneous with very obscure trace of rufous stripe in some specimens, never contrastingly striped; cubital vein obscured by dark shade; white dot at end of cell; chestnut brown shade above median shade; postmedial line indicated by dots at veins; two black dots below origin of vein Cu2; usual apical shade, accentuated by pale area extending from end of cell to apex; veins not contrasting white but lighter than membranous area; terminal dots present; fringe concolorous with ground. Hindwing dull white with infuscation increasing towards margin, more so in female. Underside forewing pinkish with central area darker; postmedial line indicated by dots at veins. Underside hindwing pinkish at costa, blending to white toward anal region. Abdomen grayish tan without dorsal tufts; caudal tuft concolorous with rest of abdomen; coremata not developed. Sexes similar but female somewhat larger. Male Genitalia (Fig. 5A): uncus, tegumen, and vinculum unmodified; valve unmodified except sensory plate displaced laterally, produced into projection; clasper modified as blunt projection. Ae- deagus (Fig. 5B) with dense row of cornuti along axis of distal half of vesica; large cornutus at distal end. Female Genitalia (Fig. 13C): ductus bursae of moderate length, opening jointly into appendix bursae and corpus bursae. Distribution. — This species has a completely circum-Caribbean distribution. Flight Period. — Beginning in April, this species has been collected in all months except October. Discussion. — Leucania microsticha Hampson, 1905 is superficially similar to L. lobrega but apparently does not enter our area. Leucania lobrega is sympatric with L. rawlinsi over much of its range and has been frequently confused with it. Primary Type Data. — Holotype male (CMNH). Verbatim label data: “DOMIN- ICAN REPUBLIC: Pedernales. 26 km N Cabo Rojo, 18-06N, 71-38W, 730 m, 16 July 1992” [white label, printed] / “C. Young, R. Davidson, S. Thompson, J. Rawlins. Mesic deciduous forest with scattered pines” [white label, printed] / “HOLOTYPE Leucania lobrega Adams” [red paper, printed]. Paratypes. — (19 3, 15 ?). CUBA. Holguin: Cayo Mambi, 1 9 (ACC). Santiago de Cuba: Sierra Maestra, 305 m, 1 3 (CMNH). DOMINICAN REPUBLIC. Monsenor Nouel: [boundary with La Vega] Loma El Casabito, sum- mit, 19°03'N, 70°3FW, 1390 m, 1 9 (CMNH). GUADELOUPE. Petit-Bourg, Route forestiere de Grosse-Montagne, 1 3 (MSA); Casse Montagne, Le Lamentin, 1 3 (MSA); St. Francois, Courcelles, ex larva, 1 9 (MSA). JAMAICA. Manchester Parish: Mandeville, 2 3 (AMNH). Portland Parish: Port Antonio, 2 3 (AMNH). St. Andrew Parish; Constant Spring, 180 m, 1 9 (AMNH); Salt Hill Road, 5 3, 1 9 (CMNH). St. Ann Parish: Moneague, 1 3 (CMNH); 3.2 mi NE Kellits near Pedro River, 490 m, 1 3 (USNM). St. Elizabeth: Balaclava, 13,19 (AMNH). 2001 Adams — Antillean Moths of the Genus Leucania 191 Fig. 4. — Male genitalia of Antillean Leucania: A. Genitalia with aedeagus removed, L. subpimctata, USA, Louisiana, St. Tammany Parish, 6 km NE Abita Springs, Genitalia Slide 2098 (MSA); B. Everted aedeagus, L. subpunctata, same data as Fig. A; C. Genitalia with aedeagus removed, L. rawlinsi, Jamaica, St. Andrew Parish, Irishtown, Genitalia Slide 2033 (MSA); D. Everted aedeagus, L. rawlinsi, same data as Fig. C. All figures at same magnification. PUERTO RICO. Mona Island, I 6, I 9 (AMNH); Rio Piedras, I S, I 9 (USNM); Toa Baja, 1 $ (USNM); no specific locality, 16,49 (USNM). ST. VINCENT. Orange Grove, 16,29 (USNM). Leucania sonroja, new species (Fig. ID, 5C-D, 13D) Diagnosis. — This is the only rust-colored Leucania in the Antilles, and as it is endemic to the Dominican Republic, it should not be misidentified. Description. — (Fig. ID). Head: palpi tan; front rust. Thorax rust with two black patagial bands; posterior portion of patagia concolorous with middorsal tuft; tegulae rust with occasional black-tipped white scales; disk rust. Forewing: 15 mm (14-15 mm), ground color rust, rather uniform, except for chestnut shade above cell; veins white; postmedial line marked by black dots on veins; fringe con- colorous with ground. Hindwing dull white, with brown scales on veins and marginal infuscation. Underside forewing pinkish, darker near costa. Underside hindwing pinkish white. Abdomen pale rust, with slight dorsal tuft on second segment; caudal tuft concolorous with rest of abdomen; coremata developed. Sexes similar. Male Genitalia (Fig. 5C): uncus, tegumen, vinculum unmodified; valve unmodified. Aedeagus (Fig. 5D) with vesica unadorned at base with long recurved appendix at mid-length; single long recurved cornutus at primary gonopore, preceded by patch of smaller, thinner recurved cornuti. Female Genitalia (Fig. 13D): ductus bursae of moderate length continued into short appendix bursae beyond corpus bursae. Distribution. — Endemic to Dominican Republic. 192 Annals of Carnegie Museum VOL. 70 Flight Period. — This species has only been observed in July, September and October, probably due to lack of collecting in its restricted range. Discussion. — The moth is distributed rather widely at low elevation in mesic pine forest in the Dominican Republic. It is not common (J. E. Rawlins, personal communication). I have not identified the affinities of this moth. This species and L. lamisma new species stand out from the other taxa covered in the treatment because of their rust red coloration. J. E. Rawlins (personal communication) has noted that they occur together on the intensely red bauxite soils of the Dominican Republic. This adaptation is seen in several endemic moths of this region, in= eluding an undescribed Faronta. Primary Type Data. — Holotype male (CMNH). Verbatim label data: ‘'DOMIN- ICAN REPUBLIC: Pedernales. 26 km N Cabo Rojo, 18-06N, 7L38W, 730 m, 16 July 1992” [white label, printed] / “C. Young, R. Davidson, S. Thompson, J. Rawlins. Mesic deciduous forest with scattered pines” [white label, printed] / “HOLOTYPE Leucania sonroja Adams” [red paper, printed]. Paratypes.—{2\ 3, 14 ?). DOMINICAN REPUBLIC. Pedernales: 23.5 km N Cabo Rojo, 18°06'N, 71°38'W, 540 m, 2 3, 2 $ (CMNH, MSA); 26 km N Cabo Rojo, 760 m, 4 3 (CMNH); 26 km N Cabo Rojo, 18°06'N, 71°38'W, 730 m, 2 3, 3 9 (CMNH, MSA); 30 km N Cabo Rojo, 18°07'N, 71°39'W, 1,070 m, 3 3, 3 9 (CMNH); 30 km N Cabo Rojo, 18°07'N, 71°39'W, 1,060 m, 1 3 (CMNH); 37 km N Cabo Rojo, 18°09'N, 71°35'W, 1,480 m, 3 3, 3 9 (CMNH); 37 km N Cabo Rojo, 4 km E La Abeja, 18°10'N, 71°37'W, 1,440 m, 1 3, 1 9 (CMNH, MSA); La Abeja, 38 km NNW Cabo Rojo, 18°19'N, 71°38'W, 1,250 m, 5 3, 2 9 (CMNH, MSA). Leucania inconspicua Herrich-Schaffer (Eig. IE, 6A-B, 13E) Leucania inconspicua Herrich-Schaffer, 1868:148. Gundlach, 1881:301; Moschler, 1890:141; Gun- dlach, 1891:172; Poole, 1989:580. Holotype male (ACC, Havana), [photograph examined]. Cirphis inconspicua (Herrich-Schaffer): Hampson, 1905:554, pi. 94, fig. 16; Dyar, 1914:176; Draudt, 1924:167, pi. 24, row I; Wolcott, 1936:161; Schaus, 1940:187. Diagnosis. — The only conspicuous marking on the forewing is a dark shade between the end of the cell and postmedial line. There should be no confusion with any other Leucania in its range. The valves of the male genitalia have char- acteristic sharp “shoulders”. Description. — (Pig. IE). Head: palpi tan with black flecks; front tan. Thorax tan with four bands on patagia; tuft of dark scales immediately behind patagia; tegulae with faint row of dark scales; disk tan. Forewing: 15 mm (15-16 mm), ground straw colored with diffuse flecks of black scales; black dot at end of cell and dark shade between end of cell and postmedial line; only conspicuous mark a small dot at origin of vein Cu2; postmedial line marked with small dots at veins; usual apical shade faint; veins somewhat lighter than ground but not conspicuous; fringe concolorous with ground. Hindwing white with fuscous shading on veins towards margin. Underside forewing tan, darker in subcostal area; postmedial line marked by faint dots at veins. Underside hindwing white with tan costa. Abdomen tan with dark tufts on first three segments; caudal tuft concolorous with rest of abdomen; coremata developed. Sexes similar. Male Genitalia (Fig. 6A): uncus, tegumen, and vinculum unmodified; valve with cucullus knob- like; valve produced into shaip shoulder at level of valvula, sensory area small; ampulla, digitus and editum unmodified; clasper produced into club-like projection at base; clavus not modified. Aedeagus (Fig. 6B) with proximal one third of vesica unornamented; vesica widens at distal two thirds at junction with low appendix bearing cluster of ten long cornuti; row of strong cornuti on opposite side, con- tinuing with increasing density to distal end, terminating in one large cornutus and tight cluster of six long spines. Female Genitalia (Fig. 13E): ductus bursae of moderate length, continued beyond corpus bursae into short, broad appendix bursae. Distribution. — The species flies from northeastern South America, throughout 2001 Adams — Antillean Moths of the Genus Leucania 193 Central America and the Antilles. It strays to extreme southern Florida (Dickel, 1991). Flight Period. — This moth undoubtedly flies throughout the year in the Antilles and has been collected in every month except February. Taxonomic notes. — Herrich- Schaffer (1868) in the original description and Gundlach (1881) both note that the number 780 is associated with the type of L. inconspicua. However there is an apparent error in the labeling of the type of this species with the number 783. The number 783 does not occur in either of these publications. Primary Type Dato.— Holotype male (ACC). Verbatim label data: “783” [small white, printed] / '"783 Leucania inconspicua HS” [large white label with black border, script; two labels glued together, only the larger with pin hole]. Material Examined. — (212 3, 125 $). CUBA. Cienfuegos: near Paso Caballos, 6 km S Cienfuegos, 10 m, 3 S (USNM). Guantanamo: Loma El Gato, 1 3 (ACC). La Habana: Santiago de Las Vegas, 1 3 (ACC). Las Tunas; Hormiguero, 1 9 (CMNH). Matanzas: Allende, 1 3 (ACC); Playa, 1 $ (ACC); Versalles [northern suburb of Matanzas], 1 $ (ACC); no specific locality, 1 3, 1 9 (USNM). Santiago de Cuba: Cuabitas [northeastern suburb of Santiago de Cuba], 5 3, 4 9 (ACC); Sierra Maestra, 305 m, 2 3, 2 9 (ANSP); no specific locality, 14 3, 6 9 (AMNH, CMNH, USNM). Subregion unknown: 2 3,5 9 (AMNH, USNM). DOMINICA. St. Joseph: Grande Savane, 5 9 (USNM). St. Paul: Clarke Hall, 11 3,4 9 (USNM); 0.4 mi E Pont Casse, 1 9 (USNM); Silvania, 1 9 (USNM); Springfield Estate, 2.5 km ENE Canefield, 15°2rN, 61°22'W, 450 m, 3 3 (CMNH); S. Chiltern, 2 3, 2 9 (USNM); Hatton Garden, 1 9 (AMNH). DOMINICAN REPUBLIC. Azua: East side of crest. Sierra Martin Garcia, 7 km WNW Barrero, 18°2rN, 70°58'W, 860 m., 5 3, 4 9 (CMNH). Barahona: 6 km NW Paraiso, Rio Nizao, 18°02'N, 71°12'W, 170 m, 2 3 (CMNH); 4 km NE Polo, 1,260 m, 1 3 (CMNH); 5 km SE Polo, slopes of Loma La Torre, 18°03'N, 71°16'W, 980 m, 2 3 (CMNH); near Barahona, 244 m, 1 9 (USNM). Baoruco: Sierra de Neiba, Los Guineos, upper Rio Colorado, 18°35'N, 71°1 1 'W, 630 m, 2 3 (CMNH). Distrito Nacional: Santo Domingo, 1 9 (USNM). Elias Pina: North slope Sierra de Neiba 2 km SW Canada, 7 km WSW Hondo Valle, 18°42'N, 71°45'W, 980 m, 2 3, 1 9 (CMNH). El Seibo: Loma Cocuyo, 6 km N Pedro Sanchez, 18°55'N, 69°07'W, 475 m, 1 3, 2 9 (CMNH). Independenda: 3 km ESE El Aguacate, north slope Sierra de Baoruco, 18°18'N, 71°42'W, 1,980 m, 1 3 (CMNH); 4 km S Los Pinos, Loma de Vientos, 18°35'N, 71°46'W, 455 m, 1 3, 1 9 (CMNH). La Altagrada: 2 km N Bayahibe, 18°23'N, 68°5LW, 10 m, 1 3 (CMNH). La Vega: Near mouth of Arroyo Los Dajaos, 5 km E Manabao, 19°04'N, 70°45'W, 740 m, 2 3 (CMNH); Constanza, Hotel Nuevo Suiza, 1,164 m, 5 3, 4 9 (USNM); 18 km SE Constanza, 18°46'N, 70°39'W, 2,310 m, 1 9 (CMNH); 23 km SE Constanza, 2225 m, 18°45'N, 70°37'W, 2 3 (CMNH); 24 km SE Constanza, 18°44'N, 70°36'W, 2, 220 m, 1 3 (CMNH); Convento, 12 km S Constanza, 13,19 (USNM); 10 km NE Jarabacoa, Hotel Montana, 520 m, 1 9 (USNM). Monsenor Nouel: [boundary with La Vega], Loma El Casabito, summit, 19°03'N, 70°31'W, 1390 m, 7 3, 9 9 (CMNH). Monte Cristi: 5 km NNE Botoncillo, 19°46'N, 71°24'W, 50 m., 3 3, 2 9 (CMNH). Pedernales; 23.5 km N Cabo Rojo, 18°06'N, 71°38'W, 540 m, 6 3, 4 9 (CMNH); 26 km N Cabo Rojo, 18°06'N, 71°38'W, 730 m, 7 3, 4 9 (CMNH); 37 km N Cabo Rojo, 18°09'N, 71°35'W, 1,480 m, 6 3, 1 9 (CMNH). Peravia: 3 km SW La Nuez, upper Rio Las Cuevas, 18°39'N, 70°36'W, 1,880 m, 1 3,2 9 (CMNH). Puerto Plata: Pico El Murazo, north slope near summit, 19°4LN, 70°57'W, 910 m, 37 3, 14 9 (CMNH); Sosua, 0 m, 1 9 (AMNH). San Juan: 8 km NE Vallejuelo, 18°42'N, 71°16'W, 690 m, 1 9 (CMNH). Santiago: 1 km NE San Jose de Las Matas, 19°2LN, 70°56'W, 540 m, 3 3 (CMNH). GRENADA. Grand Etang, 13,19 (USNM). GUADELOUPE. Petit-Bourg, Hauteurs la Lezarde, 13,19 (MSA). HAITI. Departement du Quest: Petionville, 1 9 (CUIC). Departement du Sud: Ville Eormon, 31 km NW Les Cayes, S slope Morne Eormon, Massif de La Hotte, 18°20'N, 74°0LW, 1405 m, 9 3, 2 9 (CMNH). JAMAICA. Clarendon Parish; Cumberland District, 910 m, 1 9 (AMNH); Mason River Station, 4 mi NW Kellits, 670 m, 1 3 (USNM). Manchester Parish: Mandeville, 680 m, 7 3, 4 9 (AMNH, CMNH); Newport, 1 3 (USNM). Portland Parish: Hardware Gap, 1 3, 1 9 (CMNH); 1 mi N Hardware Gap, 3 3,2 9 (USNM); Port Antonio, 1 3 (AMNH); Silver Hill Gap, 13,19 (MSA). St. Andrew Parish: Constant Spring, 1 9 (CMNH); Irishtown, 732 m, 6 3, 8 9 (MSA, USNM); 194 Annals of Carnegie Museum VOL. 70 Fig. 5. — Male genitalia of Antillean Leucania: A. Genitalia with aedeagus removed, L. lobrega, PUERTO RICO, Rio Piedras, Genitalia Slide 43,248 (USNM); B. Everted aedeagus, L. lohrega, same data as Fig, A; C. Genitalia with aedeagus removed, L. sonroja, DOMINICAN REPUBLIC, Peder- nales. La Abeja, 38 km NNW Cabo Rojo, Genitalia Slide 2746 (MSA); D. Everted aedeagus, L. sonroja, same data as Fig. C. All figures at same magnification. Kings House Est., 180 m, 1 $ (AMNH); Salt Hill, 6 d, 2 9 (CMNH); no specific locality, 180 m, 1 d (USNM). St. Ann Parish: Moneague, 14 d, 2 $ (AMNH, CMNH). St. Catherines Parish: Hel- Ishire Hills, 30 m, 1 d, 1 $ (MSA); Old Harbor, 5 d (CMNH). St. Elizabeth Parish: Santa Cruz, 1 d (CMNH). St. Thomas Parish: Blue Mountains, 1 d (CMNH); Hill Gardens, 1 d (CMNH); May Hill, 1 $ (CMNH). Trelawny Parish: near Troy, 1 d (AMNH). Parish unknown: no specific locality, 8 d, 5 $ (AMNH, CMNH, USNM). PUERTO RICO. Banos de Coamo, 18°03'N, 66°22'W, 1 d, 4 $ (MSA). ST. LUCIA. Anse la Raye: Anse Galet, 1 km SSW Anse la Raye, 13°56'N, 61°03'W, 50 m, 1 9 (CMNH). Soufriere: 1 mile NE Soufriere, 1 d (USNM). ST. VINCENT. Montreal, 2 d (USNM); Orange Grove, 1 d (USNM). Leucania chejela (Schaus) (Fig. 2A, 6C-D, 13F) Cirphis chejela Schaus, 1921:360. Draudt, 1924:163, pi. 24, row A; Schaus, 1940:185. Holotype male (USNM), Guatemala [examined]. Leucania chejela (Schaus): Poole, 1989:577. Diagnosis. — This species stands apart from the other Leucania of the Neotrop- ics with its uniquely simple female genitalia and curiously modified claval area of the sacculus of the male genitalia. The elongate forewing of the moth gives it a distinctive appearance. Description. — (Fig. 2A). Head: palpi tan with brown scales interspersed; front tan. Thorax tan; 2001 Adams — Antillean Moths of the Genus Leucania 195 patagia with three transverse dark bands, posterior portion brown; tuft of brown scales immediately behind patagia; tegulae tan with few dark scales medially; disk tan. Forewing; 17 mm (16-18 mm), somewhat more elongate than congeners; ground brown; veins white; cubital vein with brown shade below not well developed; black dot at end of cell; postmedial line indicated by dots at veins Ml and Cu2; black dot below origin of Cu2, usual apical shade accentuated by light area anteriorly; fringe concolorous with ground. Hindwing white (that of female with some marginal infuscation). Underside forewing tan, somewhat darker in subcostal area; light brown scaling on veins. Underside hindwing white with tan scales at costal margin. Abdomen light tan; small tuft of dark scales on first segment; caudal tuft concolorous with rest of abdomen; coremata well developed. Male Genitalia (Fig. 6C): uncus, tegumen, and vinculum unmodified; valve with cucullus not nar- rowed at valvulus; sensory patch, ampulla, editum, and digitus typical of the genus; clasper reduced to narrow spur; large claval projection extending above valvula. Aedeagus (Fig. 6D): row of cornuti on vesica, longest proximally, diminishing distally. Female Genitalia (Fig. 13F): unique among Antillean species; ductus bursae ending in sack-like corpus bursae; appendix bursae reduced to small protuberance on corpus bursae, terminating in ductus seminalis. Distribution. — This moth occurs throughout the Antilles and Central America. Flight Period.— species has not been collected from February to May although it flies throughout the rest of the year. Discussion. — The affinities of this moth may be in the Old World. Primary Type Data. — Holotype male (USNM). Verbatim label data: “Aug.” [printed] / “Chejel Guat” [printed] / “Schaus and Barnes coll” [printed] / “Type No. 23381 U.S.N.M.” [red printed label with hand written number] / “Cirphis chejela type Schs” [hand written label]. Material Examined. — (14 3, 15 $). CUBA. Santiago de Cuba: no specific locality, 5 3, 5 $ (AMNH, USNM). DOMINICA. No specific locality, 13,1$ (AMNH). DOMINICAN REPUBLIC. Peravia: 3 km SW La Nuez, upper Rio Las Cuevas, 18°39'N, 70°36'W, 1,880 m, 1 3 (CMNH). GUADELOUPE. Petit-Bourg, Hauteurs la Lezarde, 13,1$ (MSA). JAMAICA. Clarendon Parish: Mason River Station, 6 km NW Kellits, 671 m, 1 $ (USNM). St. Andrew Parish: Constant Spring, 13,1$ (CMNH); Irishtown, 732 m, 2 3 (MSA, USNM). St. Ann Parish: Moneague [as Monigue], 2 3 (CMNH). Parish unknown: 13,3$ (AMNH, CMNH, USNM). PUERTO RICO. Banos de Coamo, 18°03'N, 66°22'W, 13,2$ (CUIC, MSA). Leucania senescens Moschler (Fig. IF, 7A-B, 14A) Leucania senescens Moschler, 1890:142. Gundlach, 1891:172; Poole, 1989:586. Lectotype male (ZMHU), Puerto Rico [genitalia preparation and photograph examined]. Cirphis latiuscula: Wolcott, 1923:161; Draudt, 1924:167 (in part). Schaus, 1940:186. [Misidentifica- tion]. Diagnosis. — This rather nondescript species is identified by the darker shade above the cubital vein which stands out against the uniform lighter tan background of the fore wing. The female genitalia are remarkable for the long and coiled appendix bursa. Description. — (Fig. IF). Head; palpi tan with black flecks; front tan. Thorax tan with three dark patagial bands; small tuft of dark scales immediately behind patagia; tegulae tan with row of fine dots on medial border, laterally becoming darker, blending into median shade of fore wing; disk tan. Fore- wing: 15 mm (14-15 mm), ground tan; cubital vein white with light brown shade above and below; black dot at end of cell and at origin of vein Cu2; usual apical shade; postmedial line indicated by black dots at veins; veins white with tan scaling on membranous portion; fringe concolorous with ground; terminal dots present. Hindwing light tan with pearly iridescence; brown scaling on veins with marginal infuscation (especially in females). Underside forewing tan; conspicuous dot at costa at origin of postmedial line. Underside hindwing pearly white with tan costal border. Abdomen tan 196 Annals of Carnegie Museum VOL. 70 Fig. 6. — Male genitalia of Antillean Leucania: A. Genitalia with aedeagus removed, L. inconspicua, JAMAICA, Portland Parish, Silver Hill Gap, Genitalia Slide 2016 (MSA); B. Everted aedeagus, L. inconspicua, same data as Fig. A; C. Genitalia with aedeagus removed, L. chejela, Cuba, Santiago de Cuba, Genitalia Slide 43,272 (USNM); D. Everted aedeagus, L. chejela, same data as Fig. C. All figures at same magnification. with dorsal tuft of black scales on segments one and two; caudal tuft concolorous with rest of abdomen; coremata well developed. Sexes similar. Male Genitalia (Fig. 7A); uncus, tegumen, vinculum, and valves unmodified. Aedeagus (Fig. 7B) not everted. Female Genitalia (Fig. 14A): ductus bursae very long, narrow, continued into appendix bursae which is equally long but coiled. Distribution. — This is a common species over much if its range from north- eastern South America, throughout Central America and the Antilles. It strays to extreme southern Florida (Dickel, 1991). Flight Period. — This species flies throughout the year and has been collected in every month except February, May, and September. Primary Type Data. — The species was described by Moschler from one male and one female syntype. The two specimens are not conspecific. Lectotype male (ZMHU) by present designation. Verbatim label data: not available. Material Examined. — ^(153 d, 97 $). BARBADOS. St. Philip Parishs Sam Lord's Castle, 1 9 (MSA). Parish unknown: No specific locality, 1 9 (AMNH). 2001 Adams — Antillean Moths of the Genus Leucania 197 CUBA. Guantanamo: Loma El Gato, 3^,2$ (ACC). Matanzas: Allende, 1 9 (ACC). Pinar del Rio: San Vicente, near mouth of Cueva del Indio, Sierra de los Organos, 22°4rN, 83°42'W, 60 m, 4 9 (CMNH); Vinales, 1 9 (USNM). Santiago de Cuba: Cuabitas, 1 6 (ACC); San Luis, 1 S, 1 9 (ACC); Santiago, 1 d, 1 9 (ACC); Sierra Maestra, 305 m, 3 9 (ANSP); no specific locality, 8 d, 11 9 (AMNH, MCZ, USNM). Villa Clara: Santa Clara, San Bias, 1 d (ACC). DOMINICA. St. George: Fresh Water Lake, 1 S (USNM). St. Joseph: Grande Savane, 39 d, 2 9 (USNM). St. Paul: Clarke Hall, 11 6,6 9 (USNM); 0.4 mi E Pont Casse, ! 9 (USNM); 2.2 mi E Pont Casse, 1 9 (USNM); 2 mi NW Pont Casse, 1 6 (USNM); Springfield, 1 d, 1 9 (USNM); Springfield Estate, 2.5 km ENE Canefield, 15°2LN, 61°22'W, 450 m, 3 d (CMNH). S. Chiltern, 1 d, 1 9 (USNM); Hatton Garden, 1 d, 1 9 (AMNH). DOMINICAN REPUBLIC. Baoruco: Sierra de Neiba, Los Guineos, 18°35'N, 71°11'W, 630 m, 1 d (CMNH); 6 km NW Paraiso, Rfo Nizao, 18°02'N, 71°12'W, 170 m, 1 d, 1 9 (CMNH); 9.2 km NW Paraiso, confluence Rfo Nizao and Rfo Coltico, 18°03'N, 71°12'W, 230 m, 1 9 (CMNH). El Seibo: 15 km S Miches, 500 m, 1 d (USNM). La Vega: 5 km SSE Jarabacoa, 640 m, 1 9 (CMNH); Bamboo Hole Canyon, Rfo Baiguate, 5 km SE Jarabacoa, 580 m, 2 9 (CMNH). Monsenor Nouel: [boundary with La Vega] Loma El Casabito, 19°03'N, 70°31'W, 1390 m, 2 d, 3 9 (CMNH). Ped- ernales: 26 km N Cabo Rojo, 18°06'N, 71°38'W, 730 m, 2 d, 1 9 (CMNH). Puerto Plata: Pico El Murazo, 19°41'N, 70°57'W, 910 m, 2 d (CMNH). Samana: Sanchez, 2 d, 13 9 (AMNH). Santiago: Santiago, 19°28'N, 70°42'W, 175 m, 1 9 (CMNH). GRENADA. Grand Etang, 12 d, 4 9 (MCZ, USNM). GUADELOUPE. Petit-Bourg, Hauteurs la Lezarde, 1 d (MSA); Morne a Louis, 1 9 (MSA). JAMAICA. Manchester Parish: Mandeville, 20 d, 13 9 (AMNH, CMNH). Portland Parish: Hardware Gap, 1,280 m, 1 d (MSA); 1 mi N Hardware Gap, 2 d (USNM). St. Andrew Parish: Constant Spring, 1 9 (CMNH); Irishtown, 732 m, 2 d, 2 9 (MSA, USNM); Kingston, 1 9 (AMNH). St. Ann Parish: Claremont, 1 d (CMNH); Moneague, 7 d, 4 9 (AMNH, CMNH); Moneague, Phoenix Park, 2 d, 1 9 (AMNH); Ocho Rios, 1 9 (AMNH); St. Ann's Bay, 1 d (AMNH). St. Catherine Parish: Mt. Diablo, Hollymount, 830 m, 2 d, 1 9 (USNM). St. Elizabeth Parish: Santa Cruz, 1 9 (CMNH). St. Thomas Parish: Morant Bay, 1 9 (CMNH). Parish unknown: no specific locality, 1 d, 1 9 (CMNH, USNM). PUERTO RICO. Adjuntas, 1 9 (AMNH); Bayamon, 1 9 (USNM); Ciales, 1,160 m, 3 d (USNM); 4 mi SE Ciales, 1 d (USNM); Laguna Guajataca, Boy Scout Camp, 205 m, 1 9 (USNM); Pico del Esta, El Yunque, 1,000 m, 2 d, 1 9 (USNM); Reserva Forestal Guajataca, 360 m, 1 d (USNM); Banos de Coamo, 10 d and 9 (CUIC). Toa-Baja, 1 d, 1 9 (CUIC). Catano, 2 d (CUIC). San German, 1 d (CUIC). ST. LUCIA. Anse la Raye: Anse Galet, 1 km SSW Anse la Raye, 13°56'N, 61°03'W, 50 m, 2 d (CMNH). Soufriere: 1.5 mi. S Mt. Gimie, 1 9 (USNM). ST. VINCENT. Montreal, 2 d (USNM); Orange Grove, 3 d (USNM). Leucania dorsalis Walker (Fig. 2B, 7C-D, 14B) Leucania dorsalis Walker, 1856:98. Poole, 1989:579. Holotype male (BMNH), [Dominican Republic], Santa Domingo, [examined genitalia of holotype, slide Noct 5768]. Revised status. Leucania humidicola: Butler, 1890:658 (in part) (not Guenee, 1852:90) [Misidentification]. Cirphis humidicola: Hampson, 1905:530 (in part); Draudt, 1924:166 (in part) (not Guenee, 1852:90) [Misidentification] . Leucania phragmitidicola: Moschler, 1890:143; Gundlach, 1891:172 (not Guenee, 1852:89) [Misiden- tification]. Cirphis phragmitidicola: Wolcott, 1923:162; Schaus, 1940:186 (not Guenee, 1852:89) [Misidentifi- cation] . Diagnosis. — This is the only pale straw-colored species in the Antillean fauna, however the diagnosis can only be confirmed by dissection of the genitalia, which, in the male, reveals a unique modification of the clavus and small “rabbit-eared” cucullus of the valve. The small appendix bearing a cornutus at mid-length of the everted aedeagus is unique. The female genitalia could only be confused with L. toddi, however the degree of sclerotization of the ductus is much less. Description. — (Fig. 2B). Head: palpi tan with black flecks; front tan with dark brown band between base of antennae. Thorax tan with three patagial bands, first two narrow and brown, third wider and 198 Annals of Carnegie Museum VOL. 70 Fig. 7. — Male genitalia of Antillean Leucania: A. Genitalia with aedeagus removed, L. senescens, St. Vincent, Orange Grove, Genitalia Slide 43,377 (USNM); B. aedeagus, not everted, L. senescens, same data as Fig. A; C. Genitalia with aedeagus removed, L. dorsalis, Jamaica, St. Ann Parish, Rose Hill, Runaway Bay, Genitalia Slide 43,269 (USNM); D. Everted aedeagus L. dorsalis, same data as Fig. C. All figures at same magnification. violaceous; tuft of dark scales immediately behind patagia; tegulae and disk tan with row of fine black dots on former. Forewing: 15 mm (14-16 mm); ground straw yellow; cubital vein white; black dot at end of cell and at origin of vein Cu2; light brown shade below cubital vein; usual apical shade; postmedial line reduced to black dots on veins Ml and Cu2, a few black scales at costa; veins white; fringe concolorous with ground. Hindwing pearly white with brown scaling on veins, border fuscous; fringe pale yellow. Underside forewing tan; black scales in subcostal area; conspicuous black dot at junction of R1 and margin; some black scales at postmedial line in female. Underside hindwing cream color. Abdomen uniform light tan with single tuft of dark scales on first segment; caudal tuft conco- lorous with rest of abdomen; coremata not developed;. Sexes similar. Male Genitalia (Fig. 7C): uncus, tegumen, and vinculum unmodified; valve with cucullus small and club-shaped; sensory plate of valve, ampulla, digitus, clasper and editum unmodified; claval process of saccLilus large with sensory hairs on projections. Aedeagus (Fig. 7D) with small appendix on base of vesica terminating in single cornutus; thin row of cornuti at distal end of vesica. Female Genitalia (Fig. 14B): ductus bursae short, continuing imperceptibly into appendix bursae, which curves sharply right around ductus bursae. Distribution. — Leucania dorsalis has a circum-Caribbean distribution through- out the Antilles, northern South America, Central America, Mexico and extreme 2001 Adams — Antillean Moths of the Genus Leucania 199 southern Florida. It is replaced by its sister species L. adjuta (Grote), 1874 along the Gulf coast of the United States from Florida to Texas, Flight Period. — This common, widespread moth flies in every month of the year. Discussion. — Leucania dorsalis is a member of a species group, exemplified by the South American L. extenuata Guenee 1852, that share distinctive features of the female genitalia. In the “extenuata” complex (L. dorsalis Walker, 1856, L. tayaudi Guenee, 1852, L. adjuta Grote, 1874, L. infatuans Franclemont, 1972 and L. toddi new species) the appendix bursae arises ventrally from the distal end of the ductus bursae and partially encircles the ductus to the right as it extends dorsally. Taxonomic Notes. — Butler (1890) treated Leucania dorsalis Walker as a syn- onym of L. humidicola Guenee, 1852. Although L. humidicola shows a superficial resemblance to L. dorsalis, it is not closely related. Moschler (1890:143) mistook L. dorsalis for L. phragmitidicola Guenee, 1852, a North American species that does not occur south of the middle Atlantic states. Butler (1879:20) mentions, but does not name, a variety of L. dorsalis from Rio Tapajos, Brazil. The material upon which this unnamed variety is based has not been examined. Primary Type Data. — Holotype male (BMNH). Verbatim label data: “43. Leu- cania dorsalis” [large white label, printed] / “Noctuidae, Brit. Mus. slide No. 5768 9 ” [large blue label, printed] / “St. Domingo, Tweedie” [small white label, handwritten] / “type” [small, green, circular label]. Material Examined. — (126 3, 89 $). BAHAMAS. Bimini: no specific locality, 1 $ (AMNH). Great Exuma: Simons Point, 23°3rN, 50°75'W, 11 8, 2 $ (MSA, TLM). Nassau: Blue Hills, 3 8, 1 ? (CMNH). BARBADOS. St. Philip Parish: San Lord’s Castle, 4 8,1$ (MSA). St, George Parish: Grove Plantation, 1 8 (MSA). CUBA. Cienfuegos: Cumanayagua, 1 8 (USNM). La Habana: Habana, 1 $ (CUIC); Jibacoa, 1 $ (ACC). Holguin: Pinares del Mayarf, 640 m, 1 $ (USNM). Matanzas: Allende, 2 $ (ACC); Cienaga Zapata, near Playa Larga, 3 8,1$ (USNM); no specific locality, 1 8 (AMNH). Pinar del Rio: Mogote dos Hermanos, 3 km W Vinales, 150 m, 1 8,1 $ (USNM); Rio Vinales, 1 8 (USNM); Sierra del Rosario, Soroa, ca. 400 m, 1 $ (USNM). Santiago de Cuba: Cuabitas, 6 8,3$ (ACC); Sierra Maestra, 305 m, 2 8, 7 $ (ANSP, CMNH); no specific locality, 4 8,6$ (AMNH, USNM). DOMINICA. 3 km NW Pont Casse, 1 $ (USNM); Grande Savanne, 11 8, 7 $ (USNM); Cabrits Swamp, 1 $ (USNM); Clarke Hall, 2 8, 2 $ (USNM); Soufriere, 152 m, 1 8, 1 $ (USNM); S Chiltern, 1 $ (USNM). DOMINICAN REPUBLIC. Azua: Sierra Martin Garcia, 7 km WNW Barrero, 18°2UN, 70°58'W, 860 m, 1 8 (CMNH); 8 km NE Padre Las Casas, Rio Las Cuevas, 18°46'N, 70°58'W, 580 m, 1 8 (CMNH). Barahona: near Barahona, 244 m, 2 $ (USNM). Dajabon: 13 km S Loma de Cabrera, ca. 400 m, 2 $ (USNM). Elias Pina: North slope Sierra de Neiba 2 m SW Canada, 7 km WSW Hondo Valle, 18°42'N, 71°45'W, 980 m, 2 8 (CMNH). Independencia: 4 km S Los Pinos, Loma de Vientos, 18°35'N, 71°46'W, 455 m, 1 8 (CMNH). La Vega: Constanza, Hotel Neuva Suiza, 1,164 m, 2 8, 2 $ (USNM). Monte Cristi: 5 km NNE Botoncillo, 19°46'N, 71°24'W, 50 m, 1 $ (CMNH). Monsenor Nouel: Loma El Casabito, summit, 19°03'N, 70°31'W, 1,390 m, 1 $ (CMNH). Pedernales: 8 km N Cabo Rojo, 30 m, 1 8 (CMNH); 1 km S Los Arroyos, 18°14'N, 71°45'W, 1,125 m, 1 $ (CMNH); Las Mercedes, 21 km N Cabo Rojo, 490 m, 1 8 (CMNH); 23.5 km N Cabo Rojo, 18°06'N, 71°38'W, 540 m, 14 8, 4 $ (CMNH); 26 km N Cabo Rojo, 18°06'N, 71°38'W, 730 m, 4 8 (CMNH); 30 km N Cabo Rojo, 18°07'N, 71°39'W, 1,070 m, 12 8, 3 $ (CMNH); 37 km N Cabo Rojo, 18°09'N, 71°35'W, 1,500 m, 1 $ (CMNH); 38 km NNW Cabo Rojo, La Abeja, 18°09'N, 7r38'W, 1,250 m, 1 $ (CMNH). Puerto Plata: Pico El Murazo, 19°4UN, 70°57'W, 910 m, 1 8, 7 $ (CMNH). Santiago: 1 km NE San Jose de Las Matas, 19°21'N, 70°56'W, 540 m, 1 8 (CMNH). San Juan: 7 km N Arroyo Cano, 1 km S Los Frios, 18°52'N, 71°01'W, 1,120 m, 1 8 (CMNH). GUADELOUPE. Petit-Bourg, Hauteurs La Lezarde, 1 $ (MSA). HAITI, Departement unknown: La Morniere, 1 $ (AMNH). 200 Annals of Carnegie Museum VOL. 70 Fig. 8. — Male genitalia of Leucania latiuscula: A. Genitalia with aedeagus removed, CUBA, Guan- tanamo, Baracoa, Genitalia Slide ELT 52 (USNM); B. Everted aedeagus, same data as Fig. A; C. Everted aedeagus, PUERTO RICO, Bayamon, Genitalia Slide 3173 (MSA); D. Everted aedeagus, DOMINICAN REPUBLIC, Puerto Plata, Pico El Murazo, Genitalia Slide CM40 (CMNH). All figures at same magnification. JAMAICA. Clarendon Parish: Portland Ridge, near Jackson Bay Cave, ca. 10 m, 1 (3 (USNM). Manchester Parish: Mandeville, 2 (3, 3 9 (CMNH); Mandeville, ca. 675 m, 2 9 (AMNH, USNM). Portland Parish: 2 km N Hardware Gap, 13,29 (USNM); Hardware Gap, 1,280 m, 1 3, 1 9 (MSA). St. Andrew Parish: Constant Spring, 1 3 (CMNH); Irishtown, 732 m, 8 3, 8 9 (MSA); Salt Hill, 2 3 (CMNH). St. Ann Parish: Browns Town, 1 3 (CMNH); Claremont, 1 9 (CMNH); Moneague, 1 3 (CMNH); Rose Hill, Runaway Bay, 274 m, 1 3,1 9 (USNM). St. Catherine Parish: Mt. Diablo Hollymount, 830 m, 2 3, 1 9 (USNM); Hellshire Hills, ca. 30 m, 1 9 (MSA); Old Harbor, 1 9 (CMNH). St. Thomas Parish: Bath, 1 9 (CMNH); May Hill, 1 9 (CMNH); Morant Bay, 1 3 (CMNH). Trelawny Parish: Baron Hill, Jackson Town, 1 3 (CMNH). Parish unknown: no specific locality, 1 3, 1 9 (CMNH). PUERTO RICO. Banos de Coamo, 18°03'N, 66°22'W, 3 3 (AMNH, MSA); Ensenada, 1 9 (AMNH); Guanica, Bosque Estatal, 50 m, 17°45'N, 66°50'W, 1 3 (MSA); Maricao, Hacienda Juanita, 18°10'N, 67°00'W, 2 3,19 (MSA); Caales [dales], 1,006 m, 1 3 (AMNH); El Senil, near Villalba, 518 m, 1 9 (USNM). ST. BARTHELEMY. Petite-Anse, 1 9 (MSA). ST. CROIX. Christiansted, 1 9 (USNM); no specific locality, 2 3,3 9 (CUIC). ST. MARTIN. Pic du Paradis, 1 3 (MSA). ST. VINCENT. Orange Grove, 1 3 (USNM). Leucania latiuscula Herrich- Schaffer (Fig. 2C, 8A-D, 14C) Leucania latiuscula Herrich-Schaffer, 1868:148. Gundlach, 1881:301; Moschler, 1890:142; Gundlach, 1891:172; Holotype male (ACC), Cuba [photograph examined]. Diagnosis. — The species may be recognized by the longitudinal shade that ex- tends from the midpoint of the cubital vein to beyond the end of the cell but not reaching the margin. 2001 Adams — Antillean Moths of the Genus Leucania 201 Description. — (Fig. 2C). Head: palpi predominantly black becoming brown dorsally; front tan with darker bars before and above base of antennae. Thorax tan with two fine transverse brown bands, followed by distinct black band on patagia; tuft of dark scales immediately behind patagia; tegulae tan with few black-tipped scales forming line medially; disk tan. Forewing: 15 mm (14-15 mm), ground tan; cubital vein surrounded by dark brown shade widening and intensifying from base of wing to slightly beyond end of cell; faint chestnut shade above dark shade; usual apical shade; small white dot at end of cell; two black dots below origin of vein Cu2, postmedial line marked by row of black dots on veins; partial second row of black dots inside postmedial line below cell; fringe brown. Hindwing white with pearly iridescence (female with slight marginal infuscation). Underside forewing tan; subcostal area darker; fringe dark; conspicuous black dot on costa at origin of postmedial line. Underside hindwing white with tan costal margin. Abdomen tan with dorsal tufts of dark scales on segments one to three; caudal tuft brown, contrasting; coremata present. Sexes similar. Male Genitalia (Fig. 8A): uncus, tegumen, and vinculum unmodified; valve unmodified. Aedeagus (Fig. 8B-D): vesica unadorned proximally; at midlength with small group of cornuti; on opposing side at midlength with two or more cornuti on small appendix (sometimes divided into two smaller appendices, each bearing one cornutus); unarmed segment follows, then short linear array of strong cornuti in diminishing row to single cornutus at terminus. Female Genitalia (Fig. 14C): ductus bursae long; appendix bursae arises from corpus bursae at termination, curving sharply left and cephalad. Distribution. — This species is a member of a species complex which includes L. punctifera Moschler, 1880, a widespread species occurring in unforested por- tions of Brazil and northeastern South America, and L. jaliscana Schaus, 1898 which represents the complex throughout Central America. Leucania latiuscula occurs throughout the Greater Antilles. Flight Period. — ^This species has only been collected in July, October and No- vember. Discussion. — The female genitalia of the punctifera / jaliscana / latiuscula complex are distinctive for each species. The valves of the male genitalia, how- ever, are very similar. In all populations of L. latiuscula examined in this study, the basic structure of the aedeagus was constant, but the number of cornuti present in the two proximal groups of cornuti varied both between and among populations (Fig. 8B--D). In the extreme case, the appendix bearing one of these cornutal groups is subdivided into two smaller appendices, each bearing a single cornutus (Fig. 8B). These modifications of the everted vesica are too variable to be diag- nostic, and therefore I treat all Greater Antillean populations as L. latiuscula, distinct from the two mainland species, L. jaliscana and L. punctifera. Taxonomic Notes. — The almost universal confusion surrounding the proper identification of this species can be traced to Hampson (1905). He applied the name incorrectly to L. subpunctata Harvey. This mistake was avoidable since Herrich-Schaffer’s description of the size and forewing marking clearly differ- entiate L. latiuscula from other Leucania occurring in the Antilles. The decisive element in the description . .sie ist um Vs kleiner als die europ. comma, der dunkle Langestreif aus der Mitte der W zieht sich uber den weisen Mittelhaken bis zum Saume,” . it is Va smaller than the European comma, the dark lon- gitudinal stripe goes from the middle of the wing across the white middle hook to the border, . . . ”]. Leucania subpunctata is approximately the size of L. comma and the dark longitudinal stripe does not extend beyond the cell. Examination of high quality photographs of the holotype (taken by J. E. Rawlins, deposited CMNH) reveal the specimen to be badly deteriorated due to exposure and museum pests, however it has been possible to exclude Cuban congeners and resolve this determination. In addition to the three Antillean species, L. subpunctata, L. punctifera, and L. senescens, treated in this review, there are three Mexican species, L. tinila Schaus, 202 Annals of Carnegie Museum VOL. 70 1894, L. misteca Schaus, 1898, and L. orizaba Schaus, 1898, presently listed in the synonymy of L. latiuscula. These names are not equal to any Antillean species and will be treated in a forthcoming work on the Central and South American species. Primary Type Data. — Holotype male [ACC, Havana]. Verbatim label data: “281” [small white label, printed] / ''281 Leucania latiuscula HS” [large white label with black border, script. These two labels are glued together and only the larger bares a pin hole]. Material Examined. — (17(5, 24 $). CUBA. Guantanamo: Baracoa, 1 5 (USNM). Matanzas: no specific locality, 1 $ (USNM). DOMINICAN REPUBLIC. Azua: 8 km NE Padre Las Casas, Rfo Las Cuevas, 18°46'N, 70°53'W, 580 m, 2 5, 1 $ (CMNH); Sierra Martin Garcia, 7 km WNW Barrero, 18°21'N, 70°58'W, 860 m, 1 9 (CMNH). Barahona: 6 km NW Paraiso, Rfo Nizao, 18°02'N, 71°12'W, 170 m, 1 5 (CMNH). La Estrelleta: 4 km SE Rfo Limpio, ca. 760 m, 1 5 (USNM). La Vega: Near mouth of Arroyo Los Dajaos, 5 km E Manabao, 19°04'N, 70°45'W, 740 m, 4 $ (CMNH); [boundary with Monsenor Nouel], Loma El Casabito, summit, 19°03'N, 70°31'W, 1,390 m, 2 $ (CMNH). Monsenor Nouel: Paso Alto de Casabito, 8 km NW La Ceiba, 19°02'N, 70°29'W, 1,280 m, 1 5 (CMNH); 1 km E Paso Alto de Casabito, 7 km NW La Ceiba, 19°02'N, 70°29'W, 1,130 m, 1 5 (CMNH). Monte Cristi: 5 km NNE Botoncillo, 19°46'N, 71°24'W, 50 m, 2 5, 4 $ (CMNH). Peravia: 3 km SW La Nuez, upper Rfo Las Cuevas, 18°39'N, 70°36'W, 1,880 m, 1 $ (CMNH). Puerto Plata: Pico El Murazo, 19°4rN, 70°57'W, 910 m, 2 5,4 9 (CMNH). Samana: Samana Peninsula, 8 km S Las Galeras, Punta Balandra, 19°1 1'N, 69°14'W, 35 m, 1 9 (CMNH). JAMAICA. St. Mary Parish: Gray’s Inn, 1 5 (AMNH). PUERTO RICO. Toa-Baja, 1 5 (CUIC); Catano, Villa Margarita, 1 5 (CUIC); San Juan, 1 9 (CUIC); Banos de Coamo, 1 9 (CUIC); Bayamon, 1 5 (USNM); Laguna Guajataca, Boy Scout Camp, 205 m, 3 5,2 9 (USNM); Guanica, 1 9 (AMNH); no specific locality, 1 5 (USNM). Leucania humidicola Guenee (Fig. 2D, 9A-B, 14D) Leucania humidicola Guenee, 1852:90. Butler, 1890:658; Hampson, 1898:244; Poole, 1989:580. Ho- lotype female (MNHN), Lrench Guiana, [genitalia preparation and photograph examined. Lecto- type designation: Viette, 1951:160]. Cirphis humidicola (Guenee): Hampson, 1905:530 (in part), pi. 93, fig. 9. Cirphis multilinea: Hampson, 1905:482 (in part); Draudt, 1924:163 (in part) (not Walker, 1856:97) [Misidentification] . Diagnosis. — The males of this species can be distinguished from all other An- tillean Leucania by the heavily tufted fore and middle tibiae. The lower angle of the clasper of the male genitalia is extended into a projection which reaches the border of the valve and beyond. The female genitalia show a characteristic ap- pendix bursae located immediately below the ostium and a very long coiled ductus bursae. Description. — (Pig. 2D). Head: palpi tan with darker scaling on dorsum; front light tan. Thorax tan with three patagia bands, first with black-tipped scales, middle less distinct, posterior line very distinct; tegulae tan with a few black scales; disc tan; distinct sex tufts on fore and middle tibiae. Porewing: 16 mm (14-17 mm), ground light brown; cubital vein white with brown shade for entire length; black dot at end of cell and below origin of vein Cu2; postmedial line indicated by dots at veins, strongest at veins Ml and Cu2; usual apical shade; black dash between veins Ml and M2; veins white; terminal dots present. Hindwing pearly white with some dark scaling on veins [except females from Antigua which are heavily infuscated]. Underside forewing light tan, darker in subcostal area. Underside hindw- ing pearly white except costal margin tan. Abdomen light tan, shaggy; caudal tuft concolorous with rest of abdomen; coremata well developed. Sexes similar, except females somewhat darker. Male Genitalia (Pig. 9A): uncus, tegumen, and vinculum unmodified; valve with cucullus short and round; ampulla, digitus and editum unmodified; clasper produced at base into projection, which reaches margin of valve; slight prominence at clavus. Aedeagus (Pig. 9B) with vesica unadorned for proximal 2001 Adams — Antillean Moths of the Genus Leucania 203 Fig. 9. — Male genitalia of Antillean Leuconia: A. Genitalia with aedeagus removed, L. humidicola, USA, Florida, Monroe County, Key Largo Key, Genitalia Slide 577 (MSA); B. Everted aedeagus, L. humidicola, same data as Fig. A; C. Genitalia with aedeagus removed, L. lamisma, DOMINICAN REPUBLIC, Pedernales, 37 km N Cabo Rojo, Genitalia Slide JER 838 (CMNH); D. Everted aedeagus, L. lamisma, same data as Fig. C. All figures at same magnification. third of length, at which point a long appendix arises terminating in single cornutus; short row of cornuti occur at distal end of vesica. Female Genitalia (Fig. 14D); ductus bursae long and coiled distally, appendix bursae arises at base of ductus bursae. Distribution. — This species is distributed from northern South America throughout the Antilles. In Florida, Texas and southern California a form occurs which may be conspecific. A population on the Galapagos Islands may also be conspecific (personal communication, Dr. A. LeGrain, 1985). This species is re- placed by an undescribed sister species which ranges from Arizona through Mex- ico, Guatemala, and Costa Rica to Panama. In temperate North America, a number of related species, exemplified by L. phragmitidicola Guenee, 1852 and L. mul- tilinea Walker, 1856, share the heavily tufted fore and middle tibiae. Flight Period. — Apparently this species does not fly from May to September, however it has been collected throughout the rest of the year. 204 Annals of Carnegie Museum VOL. 70 Primary Type Data. — Holotype female (MNHN). Verbatim label data: not available. Material Examined. — (63 6, 34 9). ANTIGUA. 1 km N airport, Winthorpes Bay, 16 6', 5 $ (USNM). Wallings Res., Fig Tree Hill, 3 c3, 2 $ (USNM). Collidge 1 (3 (USNM). BAHAMAS. Bimini: no specific locality, 4 S (CMNH). Great Exuma: Simons Point, 23°31'N, 50°75'W, 5 (3, 2 $ (MSA, TLM). BARBADOS. St. Philip Parish: Sam Lord’s Castle, 3 c3, 1 ? (MSA). St. Michael: Bridgetown, 1 c3 (AMNH); no specific locality, 1 9 (USNM). DOMINICAN REPUBLIC. Azua: 8 km NE Padre Las Casas, Rfo Las Cuevas, 18°46'N, 70°53'W, 580 m, 1 9 (CMNH). GUADELOUPE. No specific locality, 1 (3 , 1 9 (MSA). HAITI. Departement du Ouest: Petionville, 2 9 (CUIC). PUERTO RICO. Guayama, Aguirre Central, 2 c3 (CUIC). Banos de Coamo, 3 <3, 3 9 (CUIC). Palmas Abajas (near Guayama), 1 9 (CUIC). ST. BARTHELEMY. Petit-Anse, I (3, 1 9 (MSA). ST. CROIX. Christiansted, 1 c3 (USNM). Orangegrove, West End, 4 c3, 5 9 (USNM). Kingshill 3 (3, 5 9 (CUIC, USNM). Rust Up Twist 5 c3, 1 9 (USNM). Mount Eagle 1 (3, 1 9 (USNM). 2 km W Airport, 4 <3 (USNM). ST. JOHN. Gallows Point, 1 c3 (USNM). ST. LUCIA. 2 km NW Soufriere, 1 c3, 1 9 (USNM). ST. VINCENT. Orange Grove, 1 6 (USNM). TURKS AND CAICOS ISLANDS. Grand Turk: Cockburn Town, 1 9 (CMNH). Providendales: Erebus Hotel, 21°48'N, 72°15'W, 2 (3 (CMNH). Leucania lamisma, new species (Fig. 2E, 9C-D, 14E) Diagnosis. — This species shares the same forewing pattern as its Cuban sister species L. secta Herrich-Schaffer, 1868. It is endemic to Hispaniola and can be diagnosed with certainty by examination of the genitalia. The cucullus of the male valve is more elongated, and the long row of cornuti on the aedeagus is coarser than L. secta. Description. — (Fig. 2E). Head: palpi tan with darker scaling dorsally; front tan. Thorax tan with single black band on patagia, posterior portion of patagia with chestnut colored scales blending into middorsal tuft; tegulae tan with occasional black scales; disk tan. Forewing: 15 mm (14-15 mm), ground color chestnut, lighter below cubital vein; cubital vein brown for proximal one third, bright white to end of cell; brown shade under cubital vein to end of cell; usual apical shade; postmedial line marked by black dots on veins; veins M3-Cul white; fringe concolorous with ground. Hindwing creamy white with brown scales on veins; infuscation near margin, darker in females. Underside forewing pinkish; fringe contrasting dark and light bands. Underside hindwing cream with pinkish tinge at costal margin. Abdomen tan without dorsal tuft; caudal tuft concolorous with rest of abdomen; coremata developed. Sexes similar. Male Genitalia (Fig. 9C): uncus, tegumen, and vinculum unmodified; valve unmodified except for small protuberance in claval area. Aedeagus (Fig. 9D): with cluster of four cornuti on proximal portion of vesica; on opposite side a dense row of cornuti begins after short unadorned segment and continues as diminishing row distally; three very long, thin cornuti at terminus. Female Genitalia (Fig. 14E): ductus bursae continued beyond corpus bursae into appendix bursae for approximately length of ductus bursae. Distribution. — This moth has a very restricted range in the Dominican Republic (J. E. Rawlins, personal communication). It occurs in high pine savannah “balds” at an altitude of 1,300-1,500 meters. However, in this habitat it is common and makes up to 80-90% of the moths collected. Flight Period. — Due to limited collecting in the restricted range of this species, it is only known from July, September and October. Primary Type Data. — Holotype male (CMNH). Verbatim label data: “DOMIN- 2001 Adams — Antillean Moths of the Genus Leucania 205 Fig. 10. — Male genitalia of Antillean Leucania: A. Genitalia with aedeagus removed, L. secta, CUBA, Santiago, Genitalia Slide ELT 1191 (USNM); B. Everted aedeagus, L. secta, same data as Eig. A; C. Genitalia with aedeagus removed, L. toddi, Cuba, Camagiiey, Florida, Finca La Ciega, Genitalia Slide ELT 1199 (USNM); D. aedeagus, not everted, L. toddi, same data as Fig. C. All figures at same magnification. ICAN REPUBLIC: Pedernales. 37 km N Cabo Rojo (18-09N, 7U35W) 1500 m. 11 July 1987 R. Davidson, J. Rawlins” [white, printed] / “HOLOTYPE Leucania lamisma Adams” [red paper, printed]. Paratypes, — (85 d, 82 $). DOMINICAN REPUBLIC. Independencia^ 3 km ESE El Aguacate, north slope Sierra de Baoruco, 18°18'N, 71°42'W, 1,980 m, 1 $ (CMNH). Pedernales: 5 km NE Los Arroyos, 18°15'N, 71°45'W, 1,680 m, 2 9 (CMNH); 8 km. NE Los Arroyos, 18°16'N, 71°44'W, 1,940 m, 2 3, 2 9 (CMNH); 23.5 km N Cabo Rojo, 18°06'N, 71°38'W, 540 m, 1 3 (CMNH); 26 km N Cabo Rojo, 18°06'N, 71°38'W, 730 m, 6 3, 7 9 (CMNH, MSA); 30 km N Cabo Rojo, 18°07'N, 71°39'W, 1,060 m, 14 3, 1 9 (CMNH); 30 km N Cabo Rojo, 18°07'N, 71°39'W, 1,070 m, 1 3; 31 km N Cabo Rojo, 18°07'N, 71°39'W, 1,200 m, 5 3, 5 9 (CMNH); 37 km N Cabo Rojo, 18°06'N, 71°35'W, 1,480 m, 53 3, 61 9 (CMNH, MSA); La Abeja, 38 km NNW Cabo Rojo, 18°09'N, 71°38'W, 1,160 m, 2 3, 1 9 (CMNH). HAITI, Departement unknown: no specific locality, 1 3 (MCZ). Leucania secta Herrich- Schaffer (Eig. 2E, lOA-B, 14E) Leucania secta Herrich-Schaffer, 1868:147. Poole, 1989:586. Type missing; label only remains (ACC, Havana) [photograph of labels examined]. Neotype designated below. 206 Annals of Carnegie Museum VOL. 70 Cirphis hampsoni Schaus, 1940:185. Holotype male (BMNH), Bahamas [illustration examined]. New Synonym. Cirphis clarescens: Hampson, 1905:519, pi. 92, fig. 21; Wolcott, 1923:161; Draudt, 1924:164, pi. 24, row A; (not Moschler, 1890:143) [Misidentifications]. Diagnosis. — This small brown moth, endemic to Cuba and the Bahamas has a distinctive longitudinal stripe on the forewing. The cubital vein is brown for the proximal one third, then bright white to the end of the cell. It is similar to its sister species L. lamisma new species, however, it is smaller and because each is endemic to a different island the two are unlikely to be confused. Description. — (Fig. 2F). Head: palpi tan with dark scaling dorsally; front tan. Thorax tan with single black band on patagia, posterior portion of patagia with chestnut colored scales blending into middorsal tuft; tegulae tan with rare black scales; disk tan. Forewing: 13 mm, ground color brown, lighter below cubital vein; cubital vein brown for proximal one third, then bright white to end of cell; brown shade under cubital vein to end of cell; usual apical shade; postmedial line marked by black dots on veins; veins M3 and Cul white; fringe concolorous with ground. Sexes similar, except female pattern more contrasting. Hindwing creamy white with brown scales on veins; infuscation near margin, darker in females. Underside forewing pinkish; fringe contrasting dark and light bands. Underside hindwing cream with pinkish tinge at costal margin. Abdomen tan without dorsal tufts; caudal tuft concolorous with rest of abdomen; coremata developed. Male Genitalia (Fig. lOA): uncus, tegumen, and vinculum unmodified; valve with base of cucullus developed into acute angle, approximately triangular in shape; sensory plate, ampulla, digitus, editum unmodified; clasper projects to margin of valve; claval area not developed, covered with sensory hairs. Aedeagus (Fig. lOB) with row of cornuti extending length of vesica; patch of cornuti at base of vesica; two long cornuti at distal end. Female Genitalia (Fig. 14F): ductus bursae of moderate length, terminating in short sclerotized area which connects corpus bursae and appendix bursae. Distribution. — Leucania secta is endemic to Cuba and the Bahamas. It appears to be rather widespread on Cuba, as evidenced by the widely scattered collecting localities. Flight Period. — This species appears to fly in June and July and again in No- vember and January. Discussion. — Leucania secta is replaced in Hispaniola by its sister species L. lamisma. Taxonomic Notes. — No type specimen of this species exists in the Gundlach Collection. There is a pin that had previously born a specimen but now bears only the typical Gundlach Collection label. The verbatim label data: “777” and “Leucania secta HS” [large white label with black border, script]. These two labels are glued together and only the large white label has a pin hole and match the published description. It must be assumed that the type was destroyed. The Herrich-Schaffer description can only apply to one of the species of Leucania known to exist in Cuba. The critical element being, “Kenntlich durch die aus dem weissen Mpunct bis Mitte weisse Medianrippe, ...” [Recognizable from the median vein which is white from the white discal dot up to the middle, . . . ]. In no other known Cuban Leucania is the median vein white over only one half of its length. Moschler (1890) incorrectly placed L. secta as a synonym of L. commoides Guenee, 1852, a species of temperate North America that does not reach the Antilles. The material Moschler had in hand may have been the species described here as L. lobrega. Hampson (1905:519, pi. 92., fig. 21) illustrated a male of this species, collected on Andros, Bahamas, and identified it as Cirphis clarescens Moschler, 1890. This 2001 Adams — Antillean Moths of the Genus Leucania 207 Fig. 11. — Male genitalia of Antillean Leucania: A. Genitalia with aedeagus removed, L. educata, CUBA, Soledad, Genitalia Slide 2037 (MSA); B, Everted aedeagus, L. educata, same data as Fig. A; C. Genitalia with aedeagus removed, L. clarescens, Puerto Rico, Coamo Springs, Genitalia Slide ELT 1192 (USNM); D. Everted aedeagus, L. clarescens, Puerto Rico, Coamo Springs, Genitalia Slide 2157 (MSA). All figures at same magnification. misidentification was recognized by Schaus (1940:185). He had collected this species in Cuba and thinking it undescribed, named it in honor of Sir George. Primary Type Data. — Neotype female, here designated (ACC). Verbatim label data: “5171 Allende. Mtz. 29“XI-51 coll. N.T.” [white label with faint blue border and grid, script] / “CZACC 7-511585” [white label with black border, script] / “Photograph 2F in M. S. Adams, 2001. A revision of the moth Genus Leucania in the Antilles” [white label, printed] / “Neotype Leucania secta Herrich- Schaffer designated M. S. Adams 1996” [label red on one side]. Material Examined. — (16 d, 11 ?). BAHAMAS. Great Exuma: Simons Point, 23.71°N, 75.47°W, 6 d, 3 9 (MSA, TLM). Nassau: Blue Hills, 1 9 (CMNH). Los Cayos: 1 d (AMNH). CUBA. Guantanamo: no specific locality, 1 d (ANSP). Matanzas: Allende, 1 9 (ACC). Santiago de Cuba: no specific locality, 6 d, 6 9 (AMNH, USNM). Villa Clara: Marimon, 2 d (ACC). 208 Annals of Carnegie Museum VOL. 70 Leucania toddi, new species (Fig. 3 A, lOC-D, 15 A) Diagnosis. — This is the only species of Leucania with the forewing pattern obscure except over veins which are white, giving a lattice-like effect. It is en- demic to Cuba. Description, — (Fig. 3A). Head: palpi tan ventrally, shading to brown dorsally; front tan. Thorax tan with single band of brown scales on patagia; tegulae tan but laterally with brown scales blending into ground of forewing; disk tan. Forewing: 13 mm, ground uniform brown; veins white; black dot at end of cell; postmedial line marked by black dots on veins; black dot below origin of vein Cu2; apical shade obscure but pale area extending from apex to end of cell conspicuous; fringe tan. Hind wing pearly white with marginal infuscation; some brown scaling on veins, accentuated at postmedial line; fringe white. Underside forewing tan with pink tinge; subcostal area brown; postmedial line faintly marked with black scales on veins. Underside hindwing pearly white. Abdomen tan, caudal tuft con- colorous with rest of abdomen; coremata not developed. Sexes similar. Male Genitalia (Fig. IOC): uncus, tegumen, and vinculum unmodified; valve unmodified. Aedeagus (Fig. lOD): not everted. Female Genitalia (Fig. 15A): ductus bursae short, continued beyond coipus bursae into appendix bursae, which curves sharply around ductus bursae to right; corpus bursae heavily sclerotized. Distribution. — This Cuban endemic is known only from the two type speci- mens, both collected in Camagiiey Province in central Cuba. Flight Period. — The species is known only from the type series which was collected in January. Discussion. — This species is unusual among the Leucania in that the fore wing is covered by distinct brown scaling, except over the veins that are white, giving the moth a lattice-like appearance. Its close relationship to the widespread L. dorsalis is especially apparent in the male genitalia, which are closely similar but exhibit constant differences in the shape of the cucullus and clasper plate. Leu- cania toddi lacks the peculiar modification of the claval area of the sacculus that characterizes L. dorsalis. The species is endemic to Cuba. Dr. Edward L. Todd first recognized this species as distinct, and it is named in his honor. Primary Type Data. — Holotype male (USNM). Verbatim label data: “Finca La Ciega, Florida, Camagiiey, Jan. 1958” [white label, printed] / “Zayas” [white label, printed] / “E. L. Todd, Nov. 1960, 1199” [white label, printed genitalia preparation label]. Paratype. — (1 9). CUBA. Camagiiey: Florida, Finca La Ciega [same data as holotype; genitalia preparation M. S. Adams 43,258]. Leucania educata, new species (Fig. 3B, llA-B, 15B) Cirphis secta: Hampson, 1905:524, pi. 92, fig. 27; Draudt, 1924:165, pi. 24, row B; (not Herrich- Schaffer, 1868:147) [Misidentifications]. Diagnosis. — This is the only sexually dimorphic species of Leucania recog- nized to date. The male ground color is yellow, whereas that of the female is brown. In the Cuban fauna it might be confused with L. secta, but it can be distinguished by the entirely white cubital vein (the proximal one third is brown in L. secta). Description. — (Fig. 3B). Head: palpi cream colored, becoming tan dorsally; front tan. Thorax tan with single transverse band of brown scales on patagia; mid-dorsal tuft of brown scales behind patagia; tegulae tan with band of brown hairs laterally; disk tan. Forewing: 13 mm (12-13 mm), ground yellow in male, brown in female; cubital vein white with brown shade under its entire length; usual apical 2001 Adams — Antillean Moths of the Genus Leucama 209 shade; black dot at end of cell; brown dash from end of ceil to postmedial line; postmedial line indicated by black dots on veins; veins beyond postmedial line tan with area between veins darker brown; fringe brown with alternating lighter areas at ends of veins. Hindwieg of male yellow with minimal marginal infuscation; of female, brown with iefuscation throughout. Underside of male uni- form yellow except fringe brown; of female, tan with darker coloring in subcostal area and lateral one third of forewing. Abdomen tan without dorsal tufts, caudal tuft concolorous with rest of abdomen; coremata developed. Male Genitalia (Fig. 11 A): uncus, tegumen, and vinculum unmodified; valve unmodified; clasper narrov/ed into club shaped projection, which approaches margin of valve. Aedeagus (Fig. IIB): dense patch of short cornuti on vesica near base; at approximately one third its length vesica becomes wider and a linear array of cornuti continues to terminus, where it widens into large patch; single large comutus at extreme. Female Genitalia (Fig. 15B): ductus bursae moderately long and narrow, continued into short broad appendix bursae beyond corpus bursae. Distribution.— Cuban endemic has been collected at several localities on the island, however it is not known if it is associated with a particular habitat. Flight Period.— Th& species has been collected in May, June and July. There are also records from October and January. DMCMy»sio«.“=-This species is related to L. secta and L. lamisma, but can be distinguished by numerous features of the male and female genitalia. Diagnostic characters include the elongated ampulla and blunt clasper plate of the male valve, and the longer sclerotized portion of the appendix bursae of the female. Taxonomic Notes. — Hampsoe (1905) figured a yellow male of this species as L. secta. The brown female, not previously associated, was confused with L. secta by Schaus in determined material in the USNM and AMNH. The sexual dimor- phism exhibited by this species is apparently unique among the Leucania. Primary Type Data. — Holotype male (USNM). Verbatim label data; “Santiago Cuba Schaus’’ [small white label, printed] / “HOLOTYPE Leucania educata Ad- ams” [label red on one surface, printed]. Material Examined. — (33 (?, 11 $). CUBA. Holgum: P. de Mayari, 6 6 (ACC, USNM). Matanzas: Matanzas, 2 S. Alleede, 1 S (ACC); Playa, 1 $ (ACC). Santiago de Cuba: Santiago, 5 d, 5 $ (USNM); San Luis, 1 d (ACC); no specific locality, 15 5,4 $ (AMNH, MCZ, USNM); Cuabitas, 3 5, 1 ? (ACC). Leucania clarescens Moschler (Fig. 3C, IIC-D, 15C) Leucania phragmitidicola. Van? [sic]: Walker, 1856:97 (not L. phragmitidicola Guenee, 1852:89) [Misidentification of Hispaniolan material from Tweedie Collection], Leucania secta: Gundlach, 1881:300 (not Herrich-Schaffer, 1868) [Misidentification], Leucania clarescens Moschler, 1890:143. Gundlach, 1891:172; Poole, 1989:578. Lectotype male (ZMHU), Puerto Rico [genitalic dissection and photograph examined]. Meliana rosea: Hampson, 1905:586 (in part), pi. 95, fig. 28 (not Moschler, 1880:389) [Misidentifi- cation] . Neleucania rosea: Draudt, 1924:170 (in part), pL 24, row M (not Moschler, 1880:389) [Misidentifi- cation]. Diagnosis.- — ■ Leucania clarescens is distinguished from other Antillean species of Leucama by a forewing less than 15 mm in length, pale pink coloration, and a prominent projection of the sacculus in the male genitalia (Fig. 1 1C). The female genitalia have a uniquely elongated ductus bursae (Fig. 15C). Description. — (Fig. 3C). Head: palpi tan with some black-tipped scales; front tan. Thorax pinkish- tan with two indistinct black bands on patagia; tegulae and disk pinkish-tan with sparse black flecking in dark specimens. Fore wing: length, 14 mm (13-15 mm); ground pink, lightly shaded with brown; cubital vein white, with black dot at end of cell; postmedial line indicated by black dots on veins; 210 Annals of Carnegie Museum VOL. 70 Fig. 12. — Male genitalia of Antillean Leuccmia: A. Genitalia with aedeagus removed, L. incognita, CUBA, Santiago, Genitalia Slide ELT 51 (USNM); B. Everted aedeagus, L. incognita, same data as Eig. A; C. Genitalia with aedeagus removed, L. infatiians, USA, Elorida, Liberty County, Apalachicola, Genitalia Slide 1627 (MSA); D. Everted aedeagus, L. infatuans, same data as Eig. C. All figures at same magnification. apical shade extending from end of cell to apex of wing. Hindwing white with dark scaling on veins becoming fuscous at border. Underside pinkish white; in dark specimens postmedial line of forewing marked by black dots on veins. Abdomen light tan; caudal tuft concolorous with rest of abdomen; coremata not developed. Sexes similar. Male Genitalia (Fig. IIC): uncus, tegumen, and vinculum unmodified; valve with cucullus large and unmodified; sensory plate of valve large; ampulla long; digitus long; clasper reduced to narrow projection for muscle attachment; editum simple; claval process of sacculus a sharp point. Aedeagus (Fig. 1 ID) slender, unmodified, tubular, with cornuti in patch at base of vesica and at extreme distal end. Female Genitalia (Fig. 15C): ductus bursae long and narrow; appendix bursae one third as long as ductus bursae. Distribution. — This species is distributed throughout Central America from Panama to Belize as well as in the Antilles, but is replaced by an unnamed species in northern South America and by L. extincta Guenee, 1852 north of the Mexican border. Flight Period. — This species flies in June and August and again from October through February. Discussion. — A group of moths, exemplified by the North American L. extincta, share with L. clarescens derived characters of the male genitalia, including dis- 2001 Adams — Antillean Moths of the Genus Leucania 211 tinctive pattern of cornuti on the aedeagus and female genitalia, indicating a close phylogenetic relationship. Taxonomic Notes. — This small reddish species was described from Puerto Rico by Moschler, as confirmed by examination of the type. Gundlach (1881) in his discussion of L. secta, apparently confused these two taxa as his reference to, “El insecto perfecto tiene en color rojizo muy claro”, [“The adult insect is a very pale red”], can only apply to L. clarescens in the Cuban fauna. The moth figured by Hampson (1905) as Meliana rosea Moschler, from Haiti, may be L. clarescens. Examination of Mbschler’s type has established that L. rosea is an unrelated rose- colored species from South America. Leucania clarescens Mbschler, 1890, was described from 3 syntypes, 1 male and 2 females. The male specimen in the Museum fiir Naturkunde an der Hum- boldt-Universitat, Berlin, is here designated the lectotype. Primary Type Data. — Lectotype male (ZMHU). Verbatim label data: “Porto- rica, Mus. Krug 87” [square green label with black border] / “type” [small pink label] / “Zool. Mus. Berlin” [white label, printed] / “Clarescens, Mschl.” [white label, printed] / “LECTOTYPE Leucania clarescens Moschler, designated by M. S. Adams, 1992” [red paper, printed]. The lectotype was dissected by J. G. Eran- clemont and mounted in balsam on a glass slide. Material Examined. — (61 3, 88 $). CUBA. La Habana: Jibacoa, 1 S (ACC); no specific locality, 1 9 (USNM). Santiago de Cuba: Cuabitas, 1 6 (ACC); Sierra Maestra, 305 m, 1 3 (ANSP); no specific locality, 8 <5, 10 9 (AMNH, USNM). DOMINICA. Cabrits Swamp, 1 3 (USNM); Grande Savanna, 4 3, 12 9 (USNM); Roseau, Roseau River, 1 d (CMNH). DOMINICAN REPUBLIC. Azua: Sierra Martin Garcia, 7 km WNW Barrero, 18°21'N, 70°58'W, 860 m, 1 d (CMNH). Barahona: 6 km NW Paraiso, Rio Nizao, 18°02'N, 71°12'W., 170 m, 1 3 (CMNH); near Barahona, 244 m, 2 3, 5 9 (USNM). Dajabon: 9 km S Loma de Cabrera, 19°2rN, 71°37'W, 620 m, 4 9 (CMNH). Independenda: 4 km S Los Pinos, Loma de Vientos, 18°35'N, 7r46'W, 455 m, 1 d (CMNH); Sierra de Neiba just south of crest, 5 km WNW Angel Feliz, 18°41'N, 71°47'W, 1,780 m, 2 d, 2 9 (CMNH). La Vega: Constanza, Hotel Neuva Suiza, 1,164 m, 1 d (USNM); Hotel Montana, 10 km NE Jarabacoa, ca. 520 m, 1 9 (USNM); no specific locality, 2 d, 1 9 (USNM). Monsenor Nouel: 1 km E Paso Alto de Casabito, 7 km NW La Ceiba, 19°02'N, 70°29'W, 1,130 m, 2 d (CMNH). Monte Cristi: 5 km NNE Botoncillo, 19°46'N, 71°24'W, 50 m, 3 9 (CMNH). Pedernales: 23.5 km N Cabo Rojo, 18°06'N, 7r38'W, 540 m, 1 9 (CMNH); 37 km N Cabo Rojo, 18°09'N, 71°35'W, 1,480 m, 2 d, 2 9 (CMNH); 30 km N Cabo Rojo, 18°07'N, 71°39'W, 1,070 m, 1 9 (CMNH). Puerto Plata: Pico El Murazo, 19°4rN, 70°57'W, 1 d, 2 9 (CMNH). JAMAICA. Clarendon Parish: Cumberland District, ca. 914 m, 1 d (AMNH). Manchester Par- ish: Mandeville, ca. 675 m, 1 d, 3 9 (AMNH). Portland Parish: Green Hills, 1 9 (USNM). St. Andrew Parish: Constant Spring, 1 9 (CMNH); Irishtown, 732 m, 7 d, 10 9 (MSA, USNM); Kingston, 1 9 (AMNH). St. Ann Parish: Moneague, 1 9 (CMNH). St. Catherines Parish: Old Harbor, 1 d, 1 9 (CMNH). St. Elizabeth Parish: Balaclava, 1 d (AMNH). St. Thomas Parish: Morant Bay, 1 9 (CMNH). Trelawny Parish: Baron Hill, 1 9 (CMNH); near Troy, 1 d (AMNH). Parish unknown: 3 d, 9 9 (AMNH, CMNH, USNM). PUERTO RICO. Banos de Coamo, 18°03'N, 66°22'W, 13 d, 14 9 (AMNH, CUIC, MSA, USNM); Ensenada, 1 d (AMNH). Leucania incognita (Barnes and McDunnough) (Fig. 3D, 12A-B, 15D) Cirphis incognita Barnes and McDunnough, 1918:99, pi. 17, figs. 6 and 9. Holotype male (USNM), Texas [examined]. Leucania incognita (Barnes and McDunnough): Franclemont and Todd, 1983:150. Cirphis cinereicollis: Hampson, 1905:539 (in part), pi. 93, fig. 18 (not Walker, 1858:1659) [Misiden- tification], Cirphis texana: Barnes and McDunnough, 1913:20, pi. 9, fig. 14 (not Morrison, 1875:211) [Misiden- tification]. 212 Annals of Carnegie Museum VOL. 70 Fig. 13. — Female genitalia of Antillean Leucania: A. L. subpunctata, USA, Texas, Hidalgo County, Santa Ana National Wildlife Refuge, Genitalia Slide 2232 (MSA); B. L. rawlinsi, CUBA, Santiago, Genitalia Slide ELT 1196 (USNM); C. L. lobrega, ST. VINCENT, Orange Grove, Genitalia Slide 43,254 (USNM); D. L. sonroja, DOMINICAN REPUBLIC, Pedernales, 38 km N Cabo Rojo, 4 km E La Abeja, Genitalia Slide CM 14 (CMNH); E. L. inconspicua, GRENADA, Grand Etang, Genitalia Slide 43,264 (USNM); E L. chejela, GUADELOUPE, Petit-Bourg, Hauteurs La Lezarde, Genitalia Slide 834 (MSA). All figures at same magnification. 2001 Adams — Antillean Moths of the Genus Leucania 213 Diagnosis. — The only small brown Leucania in the Antilles with a dark lon- gitudinal stripe extending to the border. The male genitalia have uniquely shaped cuculli, being concave laterally. Description. — (Fig. 3D). Head: palpi tan with black flecking; front tan. Thorax tan with two black bands on patagia, posterior band extending to tegulae and connecting with longitudinal shade of forewing; tuft of dark scales immediately behind patagia; tegulae and disk tan with faint row of dark scales on former. Forewing: 14 mm (13-15 mm), ground light brown with occasional dark scales at costa; conspicuous dark shade obscuring cubital vein and extending beyond cell almost to apex; white hook-shaped spot at end of cell surrounding black dot; black dot below shade at origin of vein Cu2; postmedial line indicated by black dots on veins; usual apical shade; veins white; fringe dark. Sexes similar except ground color of female browner, more opaque. Hindwing white (female with some fuscous shading at margin). Underside forewing light tan, accentuated in subcostal area. Underside hindwing white. Abdomen uniform light tan with dark tufts dorsally, most pronounced proximally, often only apparent on first segment; two dark lines on the ventral aspect; caudal tuft yellow; coremata well developed. Male Genitalia (Fig. 12A): uncus, tegumen, and vinculum unmodified; valve with cucullus “ear” shaped because of concavity of lateral border; ampulla, digitus and editum unmodified; clasper pro- duced into lateral point for muscle attachment; clavus not produced. Aedeagus (Fig. 12B) with three long cornuti at base of vesica and row of cornuti, increasing in density and size from the proximal to distal end of vesica. Female Genitalia (Fig. 15D): ductus bursae moderately long, continued into short broad appendix bursae beyond corpus bursae. Distribution. — The range of this species is centered in Central America, ex- tending to the Greater Antilles and south Florida, northward into Texas and south- ward into Venezuela. It appears to be extending its range northward. Brou (per- sonal communication, 1985) did not encounter the species until 1985, despite nightly collecting in St. Tammany Parish, Louisiana. At present this species is the commonest Leucania in that area. Specimens in the Mississippi Entomological Museum have been collected in Alabama and Mississippi since 1990 (R. L. Brown, personal communication). A similar population increase may have oc- curred in south Florida in recent years. Kimball (1965) reported this species only from the extreme southern tip of the Florida peninsula, whereas it presently occurs as far north as Gainesville (Landolt, personal communication, 1988). A pattern of intermittent expansion and contraction of the northern limit of distribution may be typical of this and other circum-Caribbean species. Flight Period. — This species has been collected throughout the year except from February to April, July and September. Taxonomic Notes. — Barnes and McDunnough (1918:99) recognized that a short series of unidentified moths from south Texas approached Hampson’s (1905:539) figure of L. cinereicollis (Walker), 1858 and Mdschler’s (1880:389) description of L. punctifera. However, in both cases they were able to separate their material and, appropriately, described the species as Cirphis incognita Barnes and Mc- Dunnough, 1918. Examination of the type in the USNM has indicated that this species is conspecific with specimens from the Antilles and Central America. Hampson (1905:539) had Antillean material of L. incognita in hand that he mis- identified as L. cinereicollis. Leucania cinereicollis does bear a close superficial resemblance to L. incognita; however, its range does not appear to extend beyond South America. Primary Type Data. — Holotype male (USNM). Verbatim label data: “Cirphis incognita Type d B $ McD” [a red-bordered label with two horizontal red lines, hand-written] / “Brownsville Texas” [printed] / “Barnes Collection” [printed in red ink] / “Photograph PI. 9 No. 14” [blue, black bordered label, printed, numbers 214 Annals of Carnegie Museum VOL. 70 9 and 14 hand-written] / “c? genitalia on slide Sept. 21, 1936 J.F.G.C. #615” [white label bordered in black, hand-written] / “Genitalia slide By USNM 37473” [green printed label bordered in black, number hand- written] . Material Examined. — (46 3, 18 ?). CUBA. Guantanamo: San Carlos Estate, 1 9 (AMNH); no specific locality, 2 9 (ANSP). La Habana: no specific locality, 3 3 (USNM). Matanzas: no specific locality, 5 3,19 (AMNH, USNM). Pinar del Rio: 7 km N Vinales, 1 3 (AMNH). Santiago de Cuba: Ciudamar [region south of Santiago de Cuba near mouth of Bahia de Santiago de Cuba], 2 3 (ACC); Cuabitas, 3 3,2 9 (ACC); no specific locality, 9 3,6 9 (AMNH, USNM); Sierra Maestra, 305 m, 6 3, 3 9 (ANSP, USNM). Subregion unknown: 1 3 (AMNH). DOMINICAN REPUBLIC. Barahona: near Barahona, 244 m, 4 3, 1 9 (USNM). Distrito Na- cional: Santo Domingo, 1 3 (USNM). Elias Pina: North slope Sierra de Neiba 2 km SW Canada, 7 km WSW Hondo Valle, 18°42'N, 7r45'W, 980 m, 1 3 (CMNH). La Vega: Constanza, Hotel Neuva Suiza, 1,164 m, 8 3,5 9 (USNM); Hotel Montana, 10 km NE Jarabacoa, ca. 520 m, 1 9 (USNM); San Lorenzo, 1 9 (AMNH). Samana: Samana Peninsula, 8 km S Las Galeras, Punta Balandra, 19°1LN, 69°14'W, 35 m, 1 9 (CMNH); Sanchez, 3 3, 1 9 (AMNH). HAITI. Departement du Quest: Petionville, 2 3 (CUIC). JAMAICA. Manchester Parish: Mandeville, ca. 645 m, 1 9 (AMNH). St. Andrew Parish: Con- stant Spring, 1 9 (CMNH); Kingston, 1 3 (AMNH). St. Ann Parish: Strong Hill, 1 3 (AMNH). St. James Parish: Rose Hall, 1 3 (AMNH). St. Thomas Parish: Morant Bay, 2 3 (USNM). West- moreland Parish: White House, 13,19 (AMNH). Subregion unknown: 1 3 (CMNH). Leucania infatuans Franclemont (Fig. 3E, 12C-D, 15E) Leucania infatuans Franclemont, 1972:143. Franclemont and Todd, 1983:150; Poole, 1989:50. Holo- type male (Franclemont Collection, Ithaca, New York), Florida [examined]. Leucania juncicola: Kimball, 1965:91, pi. 12, fig. 11 (not Guenee, 1852:83) [Misidentification]. Diagnosis. — In the Antilles this species is most similar to L. dorsalis, but it differs in the distinctly yellow tint of the forewing and even more obviously by the white ground of the hindwing. In L. dorsalis the hindwing is distinctly fuscous, and the veins have extensive dark scales. As this species enters the Antilles only in the Bahamas, there should be little opportunity for confusion. Description. — (Fig. 3E). Head: palpi light brown with black scales scattered laterally; front brown with fine lighter lines. Thorax light yellowish tan with four conspicuous lines of dark scales tipped with white on patagia; behind patagia a middorsal tuft of pinkish brown white-tipped scales; tegulae and disc yellowish light tan. Forewing: 15 mm (13-16 mm), ground light yellowish tan; moderate dark shade below cubital vein fading beyond cell; white spot at end of cell surrounding black dot; black dot below shade at origin of vein Cu2; postmedial line indicated by black dots on veins; usual apical shade; veins white; fringe concolorous with ground. Hindwing hyaline, pearly white with fus- cous terminal line and black dots between the veins; some dark scaling of veins on outer half. Un- derside fore wing tan except accentuated in subcostal area. Underside hindwing white with prominent black dots between veins in terminal area. Abdomen uniform light yellowish tan; dorsal tufts not prominent; caudal tuft concolorous with rest of abdomen; coremata not developed. Sexes similar. Male Genitalia (Fig. 12C): uncus, tegumen, and vinculum unmodified; valve with cucullus small, barely extending beyond apex of valve; sensory plate of valve well developed; ampulla, digitus, clasper, editum, claval area of sacculus not modified. Aedeagus (Fig. 12D): with two diverticulae arising proximal to midportion of the vesica, each with a “cock’s comb” of six large cornuti; extending down side of more distal diverticulum is single row of fine cornuti, which continues almost to distal third of vesica. Female Genitalia (Fig. 15E): sclerotization of ductus bursae is continued into appendix bursae, which hooks to right under ductus bursae. Distribution. — In the original description, the species was thought to be con- fined to the southern two-thirds of the Florida peninsula; however subsequent collecting by the author has revealed the species to be widely distributed in Cen- 2001 Adams — Antillean Moths of the Genus Leucania 215 Fig. 14. — Female genitalia of Antillean Leucania: A. L. senescens, JAMAICA, St. Andrew Parish, Irishtown, Genitalia Slide 3227 (MSA); B. L. dorsalis, DOMINICA, 3 km NW Pont Casse, Genitalia Slide 43,267 (USNM); C. L. latiuscula, DOMINICAN REPUBLIC, Azua, 8 km NE^ Padre Las Casas, Rio Las Cuevas, Genitalia Slide CM41 (CMNH); D. L. humidicola, ST. BARTHELEMY, Petite Anse, Genitalia Slide 805 (MSA); E. L. lamisma, DOMINICAN REPUBLIC, Pedernales, 37 km N Cabo Rojo, Genitalia Slide CM 13 (CMNH); E L. secta, BAHAMAS, Great Exuma, Simons Point, Genitalia Slide 691 (MSA). All figures at same magnification. Fig. 15. — Female genitalia of Antillean Leucania: A. L. toddi, CUBA, Camagiiey, Florida, Finca La Ciega, Genitalia Slide 43,258 (USNM); B. L. educata, CUBA, no specific locality. Genitalia Slide 43,255 (USNM); C. L. clarescens, DOMINICAN REPUBLIC, La Vega Province, Hotel Montana, Genitalia Slide 43,284 (USNM); D. L. incognita, DOMINICAN REPUBLIC, Samana, Samana Pen- insula, 8 km S Las Galeras, Punta Balandra, Genitalia Slide CM42 (CMNH); E. L. infatuans, USA, Florida, Duval County, Jacksonville, Genitalia Slide 1629 (MSA); F. L. neiba, DOMINICAN RE- PUBLIC, Puerto Plata, Pico El Murazo, north slope near summit, Genitalia Slide CM28 (CMNH). All figures at same magnification. 2001 Adams — Antillean Moths of the Genus Leucania 217 tral America from southern Texas throughout Mexico to Belize and Guatemala. It enters the Antilles only on Bimini in the Bahamas. Flight Period- — The collecting date of the Bimini specimen is not available. Taxonomic Notes. — As noted in the discussion of L. dorsalis this species is closely related to the South American species L. extenuata Guenee, 1852. In his description of L. infatuans, Franclemont recognized this relationship and illus- trated the genitalia of L. extenuata from Surinam. Unfortunately, at the time, the type of L. humidicola Guenee, 1852 had not been examined, and the species was misidentified. Subsequent study of the adult and male and female genitalia of L. humidicola clarified this confusion. Primary Type Data. — Holotype male (CUIC). Verbatim label data: “Florida, Manatee County, Oneco, 21 March, 1957, J, G. Franclemont” [white label, print- ed]. Material Examined. — (1 6). BAHAMAS. Biminii no specific locality, 1 S (AMNH). Leucania neiba, new species (Fig. 3F, 15F) Diagnosis. — No other Antillean Leucania has the extension of the white cubital vein beyond the cell to the wing margin forming a bright white line across the middle of the entire fore wing. In general appearance this species is closest to L. humidicola', however its pattern is brighter and more contrasting. Description. — (Fig. 3F). Head: palpi light tan; front light tan. Thorax light tan with a single thin transverse band of brown scales on patagia; tegulae light tan without marks; disk light tan. Forewing: 15 mm, ground light tan; cubital vein white, strong brown shade below cubital vein; usual apical shade; black dot at end of cell; white line extends beyond cell bordered by black dash above; pale area below median shade; black dash at anal angle; postmedial line not present; no terminal spots; fringe concolorous with ground. Hindwing white without infuscation or terminal spots. Underside fore wing light tan without marks. Underside hind wing white without marks. Abdomen light tan with concolorous caudal tuft. Male Genitalia: unknown. Female Genitalia (Fig. 15F): first portion of ductus bursae sclerotized, broadly funnel-shaped, lead- ing to plicate anterior portion, then to corpus bursae and appendix bursae with no distinct differentiation of corpus bursae. Distribution. — Known only from the Cordillera Septentrional and the Sierra de Neiba in the Dominican Republic. Flight Period. — The holotype was collected in August and the only other known specimen was collected in November. Primary Type Data. — Holotype female (CMNH). Verbatim label data: “DO- MINICAN REPUBLIC: Elias Pina. North slope Sierra de Neiba 2 km SW Can- ada, 7 km WSW Hondo Valle, 980 m” [white, printed] / “18-42N, 71-45W, 29 August 1995, J. Rawlins, G. Onore, R. Davidson, Eroded field on hillside” [white, printed] / “Holotype Leucania neiba Adams” [red paper, printed]. Paratype. — (1 $). DOMINICAN REPUBLIC. Puerto Plata: Pico El Murazo, north slope near summit, 19°41'N, 70°57'W, 910 m, 1 $, $ Genitalia Slide CM28, Morton S. Adams. Checklist of Antillean Leucania Leucania Ochsenheimer, 1816. Leucania subpunctata (Harvey, 1875). Leucania rawlinsi, new species. 218 Annals of Carnegie Museum VOL. 70 Leucania lobrega, new species. Leucania sonroja, new species. Leucania inconspicua Herrich- Schaffer, 1868. Leucania chejela (Schaus), 1921. Leucania senescens Moschler, 1890. Leucania dorsalis Walker, 1856. Leucania latiuscula Herrich- Schaffer, 1868. Leucania humidicola Guenee, 1852. Leucania lamisma, new species. Leucania secta Herrich- Schaffer, 1868. Cirphis hampsoni Schaus, 1940. Leucania toddi, new species. Leucania educata, new species. Leucania clarescens Moschler, 1890. Leucania incognita (Barnes and McDunnough, 1918). Leucania infatuans Franclemont, 1972. Leucania neiba, new species. Acknowledgments The late Edward L. Todd, at the USNM, initiated this study as part of his larger work on the Noctuidae of the Antilles. I acknowledge his preliminary investigations and the kindness he extended when I assumed responsibility for continuing his project. Robert Poole (USNM) critically reviewed an early version of the manuscript, and Michael Pogue provided label data from type specimens in the USNM. David Grimaldi (AMNH) and Mark Klingler (CMNH) rendered the line drawings and John E. Rawlins (CMNH) took the photographs. This genus has been a favorite of my friend and teacher. Prof. John G. Franclemont. I thank him for sharing his wisdom and enthusiasm, as well as providing bench space in his crowded laboratory for my endeavors. John E. Rawlins has edited repeated drafts of the manuscript and consulted on many taxonomic problems. Literature Cited Barnes, W., and J. H. McDunnough. 1913. Illustrations of rare and typical Lepidoptera. Contribu- tions to the Natural History of the Lepidoptera of North America, 2(1): 1-44. . 1918. Notes and new species. Contributions to the Natural History of the Lepidoptera of North America, 4(2):6 1-208. Birch, M. C. 1972. Male abdominal brush-organs in British noctuid moths and their value as a taxonomic character. Part I. The Entomologist, 105:185-205. Bruner, S. C., L. C. Scaramuzza, and A. R. Otero. 1975. Catalogo de los Insectos que Atacan a las Plantas Economicas de Cuba. Segunda Edicion Revisada y Aumentuda. Academia de Ciencias de Cuba, Institute de Zoologica, Havana, Cuba. Butler, A. G. 1879. On the Lepidoptera of the Amazons, collected by Dr. James W. H. Trail, during the years 1873 to 1875. Part. III. — Noctuites. Transactions of the Entomological Society of Lon- don, 1879:19-76. . 1 890. Further notes on the synonymy of the genera of Noctuites. Transactions of the Ento- mological Society of London, 1890:653-691. Calora, F. B. 1966. A revision of the species of the Leucania~comp\e.x occurring in the Philippines (Lepidoptera, Noctuidae, Hadeninae). Philippine Agriculturist, 50:633-728. Crumb, S. E. 1956. The Larvae of the Phalaenidae. United States Department of Agriculture, Tech- nical Bulletin, 1 135:1-356. Dickel, T. S. 1991. New records of noctuid moths from Florida (Lepidoptera: Noctuidae). Tropical Lepidoptera, 2(l):53-58. Draudt, M. 1924. In Seitz, A. Die Gross-Schmetterlinge der Erde, Stuttgart, vol. 7, pp. 163-170. Dyar, H. G. 1902 [1903], A List of North American Lepidoptera and Key to the Literature of this Order of Insects. Bulletin of the United States National Museum, 52:1-723. . 1914. Report on the Lepidoptera of the Smithsonian Biological Survey of the Panama Canal Zone. Proceedings of the United States National Museum, 47:139-350. 2001 Adams — Antillean Moths of the Genus Leucania 219 Ferguson, D. C., D. J. Hilburn, and B. Wright. 1991. The Lepidoptera of Bermuda: their food plants, biogeography, and means of dispersal. Memoirs of the Entomological Society of Canada, 158:1-105. Forbes, W. T. M. 1936. The Cirphis pseudargyria complex (Lepidoptera: Noctuidae). Journal of the New York Entomological Society, 44:239-247. . 1954. Lepidoptera of New York and Neighboring States. Part III. Noctuidae. Cornell Uni- versity Agricultural Experiment Station, Ithaca, New York, 329:1-433. Franclemont, J. G. 1951. The species of the Leucania unipuncta group, with a discussion of the generic names for the various segregates of Leucania in North America (Lepidoptera, Phalaenidae, Hadeninae). Proceedings of the Entomological Society of Washington, 53:57-85. — . 1972. Notes on species of North American Leucania with the description of a new species (Lepidoptera: Noctuidae: Hadeninae). Proceedings of the Entomological Society of Washington, 74:141-147. Franclemont, J. G., and E. L. Todd. 1983. Noctuidae. Pp. 120-159, in Check List of the Lepidoptera of America North of Mexico (R. Hodges et al., eds.). The Wedge Entomological Research Foun- dation, Washington, D. C. Godfrey, G. L. 1972. A review and reclassification of larvae of the subfamily Hadeninae (Lepidop- tera: Noctuidae) of America north of Mexico. United States Department of Agriculture, Agricul- tural Research Service, Technical Bulletin, 1450:1-265. Grote, A. R. 1874. Notes on American Lepidoptera with descriptions of twenty-one new species. Bulletin of the Buffalo Society of the Natural Sciences, 2:145-163. 1875. On North American Noctuae. Proceedings of the Academy of Natural Sciences of Philadelphia, 27:418-427. . 1882. New Checklist of North American Moths. [No publisher listed]. . 1895. List of North American Eupterotidae, Ptilodontidae, Thyrididae, Apatelidae and Agro- tidae. Abhandlungen des Naturwissenschaftlichen Vereins zu Bremen, 14:43-128. Guenee, a. 1852. Noctuelites. Tome 1. in Histoire Naturelle des Insectes. Species General des Lep- idopteres, (J. B. A. D. Boisduval and A. Guenee, eds.). Tome Cinquieme. Librairie Encylopedique de Roret, Paris, Prance. Gundlach, j. 1881. Contribucion a la Entomologia Cubana. G. Monteil, Havana, Cuba. . 1891. La Fauna Puerto-Riquena. Lepidopteros, Anales de la Sociedad Espanola de Historia Natural, 20:109-207. Hampson, G. F. 1898. The Moths of the Lesser Antilles. Transactions of the Entomological Society of London, 1898:241-260. . 1905. Catalogue of the Lepidoptera Phalaenae in the British Museum. Volume 5. Taylor and Prancis, London, United Kingdom. Harvey, L. E. 1875. On Texan Lepidoptera collected by Mr. Belfrage. Bulletin of the Buffalo Society of Natural Sciences, 3:3-16. Herrich-Schaffer, G. A. W. 1868. Die schmetterlinge der insel Cuba. Correspondenz-Blatt des zool- ogisch-mineralogischen Vereins in Regensburg, 22:113-118, 147-156. Kimball, C. P. 1965. The Lepidoptera of Plorida. An Annotated Checklist. Division of Plant Industry, Florida Department of Agriculture, Gainesville, Florida. Lafontaine, j. D. 1987. Fascicle 27.2, Noctuoidea, Noctuidae (Part), Noctuinae (Part Euxoa). Pp. 1- 237, in The Moths of America North of Mexico (R. B. Dominick, et al., eds.). The Wedge Entomological Research Foundation, Washington, D.C. McDunnough, j. H. 1938. Checklist of the Lepidoptera of Canada and the United States of America, Part I. Macrolepidoptera, Memoirs of the Southern California Academy of Sciences, 1:1-272. Morrison, H. K. 1875. List of a collection of Texas Noctuidae, with descriptions of the new species. Proceedings of the Boston Society of Natural History, 17:209-221. Moschler, H. B. 1880. Beitrage zur Schmetterlings-Fauna von Surinam, III. Verhandlungen der kaiserlich-koniglichen zoologisch-botanischen Gesellschaft in Wien, 30 (Abhandlungen):379-486. . 1886. Beitrage zur Schmetterlings-Fauna von Jamaica. Abhandlungen der Senckenbergischen Naturforschenden Gesellschaft, 14:25-84. . 1890. Die Lepidopteran-Fauna der Insel Portorico. Abhandlungen Senckenbergischen Na- turforschenden Gesellschaft, 16:69-360. Ochsenheimer, F. 1816. Die Schmetterlinge von Europa. Volume 4. Gerhard Pleischer, Leipzig, Ger- many. Poole, R. W. 1989. Lepidopterorum Catalogus (New Series). Pascicle 118. Noctuidae [in 3 parts]. E. J. Brill/Flora and Fauna Publications, New York, New York. Schaus, W. 1894. New species of Noctuidae from tropical America. Transactions of the American Entomological Society, 21:223-244. 220 Annals of Carnegie Museum VOL. 70 . 1898. New species of Noctuidae from tropical America. Journal of the New York Entomo- logical Society, 6:107-120, 138-149. . 1921. New species of Lepidoptera in the United States National Museum. Proceedings of the United States National Museum, 59:349-396. . 1940. Scientific Survey of Porto Rico and the Virgin Islands. Volume XII — Part 2. Insects of Porto Rico and the Virgin Islands — Moths of the family Noctuidae. New York Academy of Sciences, New York, New York. Smith, J. B. 1893. A Catalogue Bibliographical and Synonymical of the Species of Moths of the Lepidopterous Superfamily Noctuidae found in Boreal America. Bulletin of the United States National Museum, 44:1-424. . 1903. Contributions toward a monograph of the Lepidopterous family Noctuidae of boreal North America. A revision of the moths referred to the genus Leucania, with descriptions of new species. Proceeding of the United States National Museum, 25:159-209. Varley, G. C. 1962. A plea for a new look at Lepidoptera with special reference to the scent distributing organs of male moths. Transactions of the Society for British Entomology, 15:29-40. Viette, P. E. L. 1951. Sur quelques noctuelles decrites par Guenee. Bulletin mensuel de la Societe linneene de Lyon, 20:159-162. Walker, F. 1856. List of the Specimens of Lepidopterous Insects in the Collection of the British Museum, 9:1-252. Edward Newman, London, United Kingdom. . 1858. List of the Specimens of Lepidopterous Insects in the Collection of the British Mu- seum, 15:1521-1888. Edward Newman, London, United Kingdom. Wolcott, G. N. 1923. Insectae Portoricensis [sic]. Journal of the Department of Agriculture, Porto [sic] Rico (San Juan), 7:1-313. . 1936. Insectae Borinquensis. A revised annotated check-list of insects of Puerto Rico. Journal of Agriculture, University of Puerto Rico, 20(l):l-27. ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 3, Pp. 221-230 17 August 2001 MIDDLE EOCENE ISCHYROMYIDAE (MAMMALIA: RODENTIA) FROM THE SHANGHUANG FISSURES, SOUTHEASTERN CHINA Mary R, Dawson Curator, Section of Vertebrate Paleontology Banyue Wang^ Abstract Ischyromyid rodents from Middle Eocene fissure fillings in the Shanghuang Quarry, Jiangsu Prov- ince, include Anatoparamys crepaturus, n.g. and sp., and four other, probably allied taxa. These rodents appear to represent an endemic lineage of Asian rodents that are derived in the direction of bunodont, non-lophate cheek teeth, suggesting a preference for a diet of soft vegetable matter. Key Words: Middle Eocene, ischyromyid rodents, Asian endemism Introduction The Ischyromyidae, members of one of the oldest known and most primitive families of rodents, have a Holarctic Paleogene distribution (Wood, 1962; Mi- chaux, 1968; Black, 1968, 1971; Dawson, 1977; Korth, 1984; Ivy, 1990; Escaw guel, 1999). The family is prominent in North America, but relatively rare in both Europe and Asia. Although reports of Ischyromyidae or Paramyidae from the Paleogene of Asia are not uncommon (for example, Li, 1963, 1975; Dawson, 1964, 1968; Sahni and Khare, 1973; Sahni and Srivastava, 1977; Hussain et aL, 1978; Li et aL, 1979; Bruije et aL, 1982; Shevyreva, 1984; Qi, 1987), a number of these taxa actually belong to other families, especially within the Ctenodac- tyloidea, not to the Ischyromyidae (Dawson, 1977; Wood, 1977; Hartenberger, 1982; Dawson et aL, 1984; Tong, 1997). To date, five moderately well-defined ischyromyid species can be recognized in the Asian Eocene, Hulgana ertnia, Eoischyromys youngi, Asiomys dawsoni, Taishanomys changlensis, and Acritoparamys? wutui. Of these Hulgana is a rel- atively derived ischyromyid known by several upper and lower jaws from the Late Eocene Ulan Gochu Formation of Inner Mongolia (Dawson, 1968). Eois- chyromys is a late or latest Middle Eocene rodent recognized from one jaw col- lected in the Changxindian Formation of northern China and one from the upper red beds (?Shara Murun Formation) of Inner Mongolia (Wang et aL, 1998). Asiomys is based on three isolated teeth from the Middle Eocene Arshanto fauna of Inner Mongolia (Qi, 1987). Taishanomys and Acritoparamys? wutui are each represented by a lower jaw with molar teeth from the Wutu Formation of Shan- dong Province (Tong and Dawson, 1995). Two other named taxa, Paramys obay- liensis and Abrosomys agasma, are each known from one isolated tooth and will not be considered further here due to their sparse record (Shevyreva, 1984). The ‘ Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences, P.O. Box 643, Beijing 100044, People's Republic of China. Submitted 23 January 2001. 221 222 Annals of Carnegie Museum VOL. 70 same is true of some isolated ischyromyid teeth that have been described from the Middle and Late Eocene of Inner Mongolia (Li, 1963; Dawson, 1964). The fossiliferous fissure fillings in Shanghuang Quarry (Triassic Shangqinglong limestone), near Shanghuang village in southern Jiangsu Province, have produced a diverse record of Paleogene mammals (Qi et ah, 1991, 1996; Wang and Dawson, 1994; Qi and Beard, 1996). Among them, ischyromyid rodents were collected from four of the five fissures (IVPP localities 93006. A [fissure A], 93006. B [fis- sure B], 93006. D [fissure D], and 93006. E [fissure E]) from which collections were made by quarrying followed by screen washing. Correlation based on the mammalian assemblages suggests an Irdinmanhan to early Sharamurunian Asian Land Mammal Age (ALMA) for the fissures, with fissure 93006. B being slightly younger than fissures 93006. D and E. All of the currently known ischyromyids are represented only by isolated teeth. At least four and possibly five taxa appear to be present. Even this incomplete record is of importance, due both to the general rarity of ischyromyids in the Asian Paleogene and to insights that these rodents can provide on their phylogenetic and geographic affinities. Methods Biochronologic terminology for Paleogene localities in China follows Tong et al. (1995). Worldwide correlations are based on Berggren and Prothero (1992). Abbreviation for repository: IVPP, Institute of Vertebrate Paleontology and Pa- leoanthropology, Chinese Academy of Science. Systematic Paleontology Order Rodentia Bowdich, 1821 Family Ischyromyidae Alston, 1876 Anatoparamys crepaturus, new genus and species Holotype.— Right M2, IVPP V 11032.1. Referred specimens.— Three right dP4 (V 11032.3, V 11035.1, V 11035.2), right P4 (V 11032.2), left M, (V 1 1032.4), right M, „,2 (V 1 1032.5, V 1 1032.6, V 1 1032.16), left M, (V 1 1032.7), right dP-* (V 11032.8), left dP^ (V 11032.12), left P^ (V 11032.9), right M' (V 11032.13), two right (V 11032.14, V 11032.15), two right M-MV 1 1032. 10, V 11032.11). Diagnosis. — Medium-sized ischyromyid (Table 1) having bunodont cheek teeth with rounded cusps and contours, very little development of lophs, which, where present, are narrow; upper cheek teeth have complete, narrow protoloph, proto- cone anteroposteriorly elongate, prominent rounded metaconule; protoconule ab- sent and very little or no trace of a hypocone; P4 elongate with narrow trigonid, small trigonid basin and protoconid; lower molars rhomboidal in shape, expanded anterolingually, trigonid basin small, talonid basin wide and shallow, entoconid very small and no trace of a hypolophid. Differs from other ischyromyids in combination of the following characters: rounded cusps of all cheek teeth; anteroposteriorly elongate protocone of P^^-M-^; absence of hypolophid of lower cheek teeth; absence of protoconule and rudi- mentary or absent hypocone on upper cheek teeth; rhomboidal shape of lower molars. Localities. — Fissure D (V 1 1032); fissure E (11035). Etymology. — Greek, anatole, east, and Paramys, a well-known genus of Paleogene ischyromyid; Latin, crepatura, fissure. 2001 Dawson and Wang — Middle Eocene Ischyromyidae, Southeastern China 223 Table 1. — Measurements (in mm.) of Shanghuang Ischyromyidae (all numbers are preceded by IVPP V). Number Locus Anteroposterior Transverse Anatoparamys crepaturus 11032.8 dP'^ 3.10 3.50 11032.9 p4 3.0 3.93 11032.13 M' 3.66 4.12 11032.14 M2 3.59 4.40 11032.10 M3 3.60 3.70 11032.11 M3 3.50 3.50 11032.3 dP4 3.40 2.90 11035.1 dP4 3.50 2.95 11035.2 dP4 3.45 2.90 11032.2 P4 4.0 3.3 11032.4 M, 3.6 3.75 11032.5 M, 3.75 3.90 11032.6 M, 3.90 4.0 11032.1 M2 (holotype) 4.0 4.0 11032.7 M3 4.0 3.6 Anatoparamys sp. 11033.3 dP4 3.12 2.7 11034 P4 3.4 3.5 11033.2 M3 3.28 3.19 ischyromyid taxon 1 11037.1 dP4 3.52 3.56 11037.2 M2 3.46 3.86 ischyromyid taxon 2 11036.1 M'™2 4.30 5.50 11036.2 M2 5.30 6.70 Description. — The isolated cheek teeth by which this new taxon is known are distinctly bunodont, with rounded outlines and cusps and little development of lophs. Two small buccal roots and one strong lingual root support the upper molars. The lower permanent cheek teeth have two strong, sturdy roots, one each under the trigonid and the talonid. The trigonid root of M3 has two pulp cavities. Lingual enamel on the permanent upper teeth extends farther up the tooth shaft than the buccal enamel, suggesting some lingual hypsodonty. P'* (Fig. IB) is a transversely wide tooth. Its most striking feature is the anteroposteriorly elongated, obliquely set protocone, which fills the lingual side of the tooth. There is no hypocone. Paracone and metacone are well developed cusps, and the metaconule is also strong. The paracone is extended lingually by a short protoloph, and the metacone by a metaloph that connects to the metaconule. In this relatively little worn tooth there is no loph from metaconule to protocone. The anterior and posterior cingula are strong transverse ridges. The three M* referred here may represent both loci, the one interpreted as M‘ being transversely narrower but anteroposteriorly longer than the probable M^ (Fig. 1C). These molars have an elongate protocone with little or no trace of any hypocone swelling, a straight complete protoloph between paracone and protocone, and a distinct, rounded metaconule. In the less worn teeth, there is no con- nection between the metaconule and the protocone, but a connection may develop following greater wear. M' has a rounded mesostyle and a small, rounded cuspule buccal to the metaconule. The more worn M^ have an elongate mesostyle and lack a discrete second metaconule. One little worn (Fig. ID) and one worn specimen of M^ share with the other upper teeth an anteroposteriorly elongated protocone. The protoloph extends to the protocone. The metaconule is very small and isolated on the less worn tooth (V 11032.11), whereas in the more worn tooth (V 11032.10) it connects to an indistinct metacone via a metaloph. M^ increases in area of the occlusal surface as it wears down, so the more worn tooth appears to be the larger of the two. Both upper and lower deciduous teeth have been referred to this taxon. Association of two specimens of dP'^ (Fig. 1 A) with Anatoparamys is supported by the similarly large, elongate and oblique protocone 224 Annals of Carnegie Museum VOL. 70 Fig. 1. — Occlusal views of cheek teeth of Anatoparamys crepaturus. A. Right dP"^, IVPP V 1 1032.8. B. Left P\ IVPP V 11032.9. C. Right IVPP V 11032.14. D. Right M-\ IVPP V 11032.11. E. Right dP4, 1 1032.3. F. Right P4, IVPP V 11032.2. G. Left M„ IVPP V 11032.4. H. Right M2, IVPP V 1 1032.1, holotype. I. Left M3, IVPP V 1 1032.7. that they share with P-^. These teeth are expanded anterobuccally by a distinct cingular shelf. The metaconule is well rounded. The only known P4 (Fig. IF) is slightly broken, lacking some enamel on the lingual and anterior surfaces. It is an elongate tooth, having a long wear facet anterior to the posterior arm of the proto- conid. The protoconid is low, the metaconid is the highest cusp and is in a line anterior to the protoconid. A narrow posterior arm of the protoconid extends obliquely forward to the metaconid. The narrow, complete ectolophid lacks any trace of a mesoconid. The large hypoconid protrudes anteriorly; posterolingually it extends directly into the curved posterior cingulid. There is no trace of an entoconid, possibly due to lingual damage to the specimen, or of a hypolophid, a real absence. The talonid basin is wide and shallow. The lower molars are rhomboidal in shape, with a lingually protruding metaconid. This shape is least well developed on M, (Fig. IG), more pronounced on M2 (Fig. IH), and most prominent on M3 (Fig. II). The ectolophid of M, appears anteroposteriorly longer and transversely narrower, whereas those on the following two molars are shorter and wider, reflecting the more rounded protoconid on the latter two molars. On M|.2 the posterior arm of the protoconid extends anterolingually to a com- parable buccal ridge on the metaconid, forming a metalophid that closes posteriorly a very constricted 2001 Dawson and Wang — Middle Eocene Ischyromyidae, Southeastern China 225 trigoeid valley, which is bounded anteriorly by a well-developed anterior cingulum. On M3 there is no buccal ridge on the metaconid and the trigonid basin is open posteriorly. A distinct notch occurs on the lingual side of the molars between metaconid and entoconid. The molar entoconids are plump and low and have no sign of loph development. The hypoconid-posterior cingulid-entoconid ridge is distinctly convex posteriorly, enclosing a somewhat elongated talonid valley. A slight indication of a hypoconulid is present on the posterior cingulid of Mj but absent on M2_3. The enamel in the talonid valley of M3 is slightly wrinkled. DP4 (Fig, IE) is an elongate tooth having the anterior occlusal surface formed by a ridge that curves from the protoconid to the anterobuccal surface of the metaconid. The metaconid is a tall, prominent cusp, whereas the protoconid is low and small. A low posterior arm of the protoconid extends obliquely forward to the metaconid; a short, low lophid that extends posteriorly from the posterior arm of the protoconid parallels a similar lophid from the posterolingual edge of the metaconid. The ectolophid is low and complete between protoconid and hypoconid. The hypoconid extends without break into the posterior cingulid, v/hich is distinctly convex posteriorly. Only a tiny swelling suggests presence of an entoconid. Two dP4 from fissure E are essentially identical to the one from D and are referred to this taxon. Comparisons. — Anatoparamys is a distinctive ischyromyid in its markedly bunodont, nearly rion-lophate cheek teeth. None of the previously known Asian Eocene ischyromyids have evolved this type of dentition. The Early Eocene is- chyromyids Taishanomys and Acritoparamys ? wutui are quite primitive but the former differs from Anatoparamys in having a long molar trigonid and the latter differs from it in having a very well-developed entoconid that is separated from the posterior cingulid. There is no clear indication from either of these more primitive rodents of close phylogenetic affiliation with Anatoparamys . The upper molar structure of the Middle Eocene Asiomys, having an anteroposteriorly short protocone and a distinct hypocone, clearly differentiates it from Anatoparamys. Eoischyromys of the latest middle Eocene of northern China is very different from Anatoparamys in having an anteroposteriorly short P4 and molars with a strong posterior arm of the protoconid and a long complete hypolophid; it has been referred to the otherwise North American subfamily Ischyromyinae (Wang et ah, 1998). The Late Eocene Hulgana is similar to Anatoparamys in lacking a distinct hypocone and a hypolophid, and having a slightly rhomboidal shape of the lower molars, and a wide, shallow talonid basin. Differently derived characters of Hulgana include its anterolingually protruding protocoee, stronger lophs of the upper cheek teeth, and anteroposteriorly short P4. Although Anatoparamys and Hulgana appear to be evolving in different directions, the morphological char- acters that they share could indicate relationship of both to an ischyromyid con- siderably earlier in the Eocene. Among non- Asian ischyromyids, Anatoparamys shares some characteristics that could be indicative of relationships with the manitshine ischyromyids (Wood, 1962; Korth, 1985), which are mostly large, North American rodents having rounded cusps that are more dominant in the occlusal pattern than are the lophs. Some species of manitshines have a P4 similar to that in Anatoparamys with a small protoconid and small trigonid basin, but this morphology is also found in several other relatively primitive ischyromyids such as Franimys and Paramys copei and cannot be relied upon to support phylogenetic relationships. The man- itshines tend to have at least some development of a hypocone in the upper molars, a primitive ischyromyid feature, and a crest between the anterior side of the hypoconid and the entoconid, characters absent in Anatoparamys. The rhomboidal shape of the lower molars, found in Anatoparamys, is not strong or absent in manitshines. Anatoparamys is a rodent that seems to be in the process of simplifying the 226 Annals of Carnegie Museum VOL. 70 Fig. 2. — Occlusal views of cheek teeth of Shanghuang ischyromyids. A-C. Anatoparamys sp. A. Right dP4, IVPP V i 1033.3; B. Right P4, IVPP V 1 1034; C. Left M3, IVPP V 1 1033.2; D-E. Ischyromyidae taxon 1. D. Left dP'^, IVPP V 11037.1; E. Left M' IVPP V 11037.2; F. Ischyromyidae taxon 2. Right M2, IVPP V 11036.2. occlusal surface of its cheek teeth while increasing its bunodonty. More primitive ischyromyids usually have more lophate cheek teeth and better-developed hypo- cone, protoconule, and entoconid (Tong and Dawson, 1995; Dawson and Beard, 1996). The simple, bunodont teeth of Anatoparamys appear to be adapted for chewing soft fruits and other vegetation, Anatoparamys sp. Specimens.— Right P4 , IVPP V 1 1034; left P4, V 1 1033.1; left M3, V 1 1033.2; right dP4, V 1 1033.3. Localities.- — Fissure D (V 11033), Fissure E (11034). Description. — These four teeth are smaller than corresponding teeth assigned to Anatoparamys crepaturus (Table 1) and, while similar in having rounded cusps and weak lophs, are also morpho- logically different from that species. Their generic association seems likely, although the small number of specimens does not allow for certainty in this assignment. On the trigonid of P4 (Fig. 2B) the metaconid is a very prominent, rounded cusp, larger relative to the protoconid than in A. crepaturus. It differs also from the P4 of A. crepaturus in having the trigonid that is less elongated anteriorly and has an even smaller trigonid basin, which is little more than an open notch between protoconid and metaconid. M3 (Fig. 2C) differs from that of A. crepaturus in having a less rhomboidal shape, related to the talonid not being expanded posterobuccally. The anterior cingulid of M3 is relatively farther forward than in A. crepaturus and the posterior arm of the proto- conid is absent. Enamel in the talonid basin is slightly wrinkled. The small dP4 (Fig. 2A) that has been referred here has a trigonid very different from that in A. 2001 Dawson and Wang — Middle Eocene Ischyromyidae, Southeastern China 227 crepaturus, lacking both the anterior lophid between protoconid and metaconid and the posterior arm of the protoconid, and having a narrower trigonid basin that is open both anteriorly and posteriorly. There is a crest from the posterolingual side of the metaconid to the entoconid, which is more cuspate than in A. crepaturus. A tentative assignment to Anatoparamys sp. is based on the open talonid basin that has no trace of a hypolophid. Ischyromyidae taxon 1 Specimens.— dP\ IVPP V 11037.1; left V 11037.2. Locality. — Fissure B. Description. — Two heavily worn upper teeth from fissure B represent a taxon different from others at Shanghuang. A wider buccal than lingual side characterizes the dP"* (Fig. 2D), which has an an- teroposteriorly elongated protocone and no hypocone. The protoloph extends nearly transversely be- tween protocone and paracone, whereas the metaloph is inclined more anterolingually. A thickening on the metaloph suggests presence of a metaconule. M' ^ (Fig, 2E) is more square in outline than dP'^. Both weak lophs of M' ^ are transversely oriented; an elongated thickening on the metaloph suggests a single or double metaconule. Although different from Anatoparamys, the weak lophs and elongate protocone suggest that this poorly known taxon may be a member of the same clade of ischyromyids. Ischyromyidae taxon 2 Specimens.— mgU M' IVPP V 11036,1; right M^, IVPP V 11036.2. Locality. — -Fissure B. Description.— Two upper molars represent the largest rodent taxon from the Shanghuang fissures. IVPP V 11036.1 is so worn that the occlusal surface shows very little detail. The second tooth, IVPP 11036,2 (Fig. 2F) is slightly worn, and its wider trigon than talon suggests that it is an M^. It has rounded cusps and very weak lophs. The large protocone is anteroposteriorly elongate, occupying the entire lingual wall of the tooth, and there is no trace of a hypocone. From the paracone, the most prominent cusp, a narrow loph extends lingually to intersect a wider loph from the protocone, together forming a protoloph on which there is no trace of a protoconule. A distinct rounded mesostyle is slightly elongated transversely. The metacone, which is set well forward, contacts a large rounded metaconule, which is doubled by a rounded buccal bud that extends into the valley between the metacone and the narrow posterior cingulum. Even in this rather worn tooth, there is no loph from the larger metaconule to the protocone. Both upper molars are slightly unilaterally hypsodont. Ischyromyidae taxon 3 Specimen.— xM’ “--s, IVPP V 12667. Locality. — -Fissure A. Description. — This upper molar is close in size and morphology to of A. crepaturus except that it has the following: an anteroposteriorly wider anterior cingular shelf; a long, narrow, complete metaloph lacking any trace of a metaconule and curving anterolingually to contact the protocone; and a second loph from metacone to protocone between the metaloph and the posterior cingulum. These features are important enough to preclude assignment to the species A. crepaturus. Discussion Ischyromyids are a relatively rare part of the rodent assemblages from the Shanghuang fissures, in which they are exceeded somewhat in abundance by members of the ctenodactyloid family Yuomyidae but greatly outnumbered by the small cricetid Pappocricetodon (Wang and Dawson, 1994). Two ischyromyid 228 Annals of Carnegie Museum VOL. 70 taxa are known from fissures D-E, Anatoparamys crepaturus and Anatoparamys sp. More poorly known are one taxon from fissure A, and two from fissure B. Anatoparamys is marked by its bunodont, nearly non-lophate cheek teeth. Its occlusal pattern appears to be in the process of simplification, in a direction somewhat reminiscent of that of sciurids. The poorly represented ischyromyid taxa 1 and 2 from deposits of fissure B resemble Anatoparamys in several characters. In both, the elongated protocone and absence of a hypocone are also features of Anatoparamys. Taxon 2 further resembles Anatoparamys in roundness of cusps and weakness of lophs. It appears to be more derived than the taxa from fissures D and E in having carried the bunodonty even farther and adding a doubled metaconule to fill the posterobuccal valley. Taxon 1 appears to have developed a rather different pattern, with narrow but more complete transverse lophs. The evidence for the affinities of these taxa is slight, but relationships with Anatoparamys may be supported by the mor- phology of the lingual wall of the upper molars. Other faunal evidence as well as the seemingly more derived character of these two taxa indicate a slightly younger age for Fissure B than for fissures D and E. Fossils from fissure A are not abundant enough to establish its age relative to those of the other fissures, and the sole ischyromyid taxon from there is too poorly known to permit more than a tentative taxonomic assignment, although this taxon also has several hallmarks of the same clade as Anatoparamys (elongate proto- cone, bunodonty). The distinctive dental characters of Anatoparamys crepaturus suggest that this may represent an endemic lineage of ischyromyids, possibly reflecting provin- cialism that is also demonstrated by other parts of the Shanghuang mammalian faunas (Qi et ah, 1996). The fissures also contain indications, in the four less well-known rodent taxa, Anatoparamys sp. and three other taxa of ischyromyids, that this is a clade that had undergone some adaptive radiation in coastal southern China. In simplification of the molar patterns this clade shows similarities, prob- ably homoplasies, to the North American manitshines and to some paramyines (Korth, 1985), most of which are, however, much larger rodents as well as having a different basic molar morphology. A diet of soft vegetation for Anatoparamys and the other Shanghuang ischy- romyids is suggested by their bunodont dentition. The considerable diversity of small primates, at least five species (Beard et ah, 1994), and insectivores (Qi et ah, 1996) preserved with the ischyromyids in the fissures is evidence supporting the interpretation of the Eocene environment around the exposed limestones as being a warm, forested ecosystem. In such surroundings, soft fruits and tender leaves were probably readily available as a diet for the ischyromyids and other small herbivorous mammals. It is likely that predators, particularly birds, were responsible for a large part of the accumulations of dental and skeletal elements of the smaller vertebrates preserved in the fissure deposits. The small rodent Pappocricetodon was probably a prime target for these predators, but the larger yuomyids, which are in the size range of all but the largest ischyromyids, were also sampled with some frequency. Thus, the relative scarcity of ischyromyids in the fissure deposits probably reflects a low frequency of occurrence of the ischyromyids in the surrounding ecosystem. So far as known, ischyromyids were not dominant in Asian Eocene faunas, whereas at that time they were more diverse and abundant in North America. The dominant Asian early and middle Eocene rodents were the ctenodactyloids (Tong, 2001 Dawson and Wang — Middle Eocene Ischyromyidae, Southeastern China 229 1997; Dashzeveg and Meng, 1998; Guo et al., 2000). Somewhat later, starting in the later Middle Eocene, there began to be an increased diversity of cricetids and zapodids. Persistence of the ctenodactyloids in northern parts of Asia into the Oligocene and earlier Miocene was probably indicative of relatively dry ecosys= terns (Wang, 1997), an environment that appears to have been inimical to ischy- romyids in Asia. Acknowledgments Thanks are extended to our colleagues Qi Tao and Guo Jianwei (IVPP), Chris Beard (CMNH), Wang Qingqieg and Di Fubao (Liyang Cultural Museum), and Ross MacPhee (AMNH), for taking part in the field work in the Shanghuaeg Quarry. Malcolm C. McKenna (AMNH) provided specimens for comparative purposes. Special thanks go to the numerous patient volunteers who sorted the im- mense volume of fossiliferous Shanghuang matrix at both the IVPP and CMNH. This project was supported by grants from the U.S. National Science Foundation (grant SBR 9221231), the Chinese National Science Foundation (grant no. 49372070), and the M. Graham Netting and Edward O’Neil funds of Carnegie Museum of Natural History. Literature Cited Beard, K. C., T. Qi, M. R. Dawson, B. Wang, and C. Li. 1994. A diverse new primate fauna from middle Eocene fissure-fillings in southeastern China. Nature, 368:604-609. Berggren, W. A., AND D. R. Prothero. 1992. Eocene-Oligocene climatic and biotic evolution: an overview. Pp. 1-28, in Eocene-Oligocene climatic and biotic evolution (D. R. Prothero and W. A. Berggren, eds.). Princeton University Press, Princeton, New Jersey. Black, C. C. 1968. The Oligocene rodent Ischyromys and discussion of the family Ischyromyidae. Annals of Carnegie Museum, 39:273-305. . 1971. Paleontology and geology of the Badwater Creek area, central Wyoming. Pt. 7. Rodents of the family Ischyromyidae. Annals of Carnegie Museum, 43:179-217. Bruijn, H. de, S. T. Hussain, and J. J. M. Leinders. 1982. On some Early Eocene rodent remains from Barbara Banda, Kohat, Pakistan, and the early history of the order Rodentia. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, B, 85(3):249-258. Dashzeveg, D., and J. Meng. 1998. New Eocene ctenodactyloid rodents from the eastern Gobi Desert of Mongolia and a phylogenetic analysis of ctenodactyloids based on dental features. American Museum Novitates, 3246:1-20. Dawson, M. R. 1964. Late Eocene rodents (Mammalia) from Inner Mongolia. American Museum Novitates, 2191:1-15. 1968. Oligocene rodents (Mammalia) from East Mesa, Inner Mongolia. American Museum Novitates, 2324:1-12. . 1977. Late Eocene rodent radiations: North America, Europe and Asia. Geobios, Memoire special 1:195-209. Dawson, M. R., and K. C. Beard. 1996. New Late Paleocene rodents (Mammalia) from Big Multi Quarry, Washakie Basie, Wyoming, Palaeovertebrata, 25(2-4):301-32L Dawson, M. R., C. Li, and T. Ql 1984. Eocene ctenodactyloid rodents (Mammalia) of eastern and central Asia. Pp. 138-150, in Papers in vertebrate paleontology honoring Robert Warren Wilson (R. M. Mengel, ed.). Carnegie Museum of Natural History, Special Publication No. 9. Escarguel, G. 1999. Les rongeurs de FEocene inferieur et moyen d’Europe occidentale. Systema- tique, phylogenie, biochronologie et paleobiogeographie des niveaux-reperes MP 7 a MP 14. Palaeovertebrata, 28(2-4):89-35 1 . Guo, J., Y. Wang, and X. Yang. 2000. A new Early Eocene ctenodactyloid rodent (Rodentia, Mam- malia) and the associated mammalian fossils from Danjiangkou, Hubei. Vertebrata PalAsiatica, 38(4):303-313. Hartenberger, j. -L. 1982. A review of the Eocene rodents of Pakistan. Contributions of the Museum of Paleontology, University of Michigan, 26(2): 19-35. Hussain, S. T, H. de Bruijn, and J. M. Leinders. 1978. Middle Eocene rodents from the Kala Chitta Range (Punjab, Pakistan). Proceedings of the Koninklijke Nederlandse Akademie van Weten- schappen, B, 81(1):74-112. Ivy, L. D. 1990. Systematics of Late Paleocene and Early Eocene Rodentia (Mammalia) from the Clarks Fork Basin, Wyoming. Contributions from the Museum of Paleontology, University of Michigan, 28(2):21-70. 230 Annals of Carnegie Museum VOL. 70 Korth, W. W. 1984. Earliest Tertiary evolution and radiation of rodents in North America. Bulletin of Carnegie Museum of Natural History, 24:1-71. . 1985. The rodents Pseudotomus and Quadratomus and the content of the tribe Manitshini (Paramyinae, Ischyromyidae). Journal of Vertebrate Paleontology, 5(2): 139-1 52. Li, C. 1963. Paramyid and sciuravids from North China. Vertebrata PalAsiatica, 7(2): 15 1-160 (in Chinese, English summary). . 1975. Yuomys, a new ischyromyid rodent genus from the Upper Eocene of North China. Vertebrata PalAsiatica, 13(l):58-70 (in Chinese, English summary). Li, C., C. Chiu, D. Yan, and S. Hsieh. 1979. Notes on some Early Eocene mammalian fossils of Hengtung, Hunan. Vertebrata PalAsiatica, 17(l):71-80 (in Chinese, English summary). Michaux, J. 1968. Les Paramyidae (Rodentia) de TEocene inferieur du Bassin de Paris. Palaeover- tebrata, 1(4): 135-193. Qi, T. 1987. The Middle Eocene Arshanto fauna (Mammalia) of Inner Mongolia. Annals of Carnegie Museum, 56:1-73. Qi, T, AND K. C. Beard. 1996. Nanotitan shanghuangensis, gen. et sp. nov.: the smallest known brontothere (Mammalia: Perissodactyla). Journal of Vertebrate Paleontology, 16(3):578-581. Qi, T, K. C. Beard, B. Wang, M. R. Dawson, J. Guo, and C. Li. 1996. The Shanghuang mammalian fauna. Middle Eocene of Jiangsu: history of discovery and significance. Vertebrata PalAsiatica, 34(3):204-214 (in Chinese, English summary). Qi, T, G. Zong, and Y. Wang. 1991. Discovery of Lushilagus and Miacis in Jiangsu and its zoo- geographical significance. Vertebrata PalAsiatica, 29(l):59-63 (in Chinese, English summary). Sahni, a., and S. K. Khare. 1973. Additional Eocene mammals from the Subathu Formation of Jammu and Kashmir. Journal of the Palaeontological Society of India, 17:31-49. Sahni, A., and M. C. Srivastava. 1977. Eocene rodents of India: their paleobiogeographic signifi- cance. Geobios, Memoire Special, 1:87-95. Shevyreva, N. S. 1984. New Early Eocene rodents from the Zaysan Basin. Pp. 77-114, in Flora i fauna Zaysanskoy vpadiny, (L. K. Gabunia, ed.). Akademiya Nauk Gruzinskoy SSR (in Russian). Tong, Y. 1997. Middle Eocene small mammals from Liguanqiao Basin of Henan Province and Yuan- qu Basin of Shanxi Province, central China. Palaeontologia Sinica, 18, C, 26:1-256 (in Chinese, English summary). Tong, Y., and M. R. Dawson. 1995. Early Eocene rodents (Mammalia) from Shandong Province, China. Annals of Carnegie Museum, 64:51-63. Tong, Y, S. Zheng, and Z. Qiu. 1995. Cenozoic mammal ages of China. Vertebrata PalAsiatica, 33(4):290-314 (in Chinese, English summary). Wang, B. 1997. The mid-Tertiary Ctenodactylidae (Rodentia, Mammalia) of eastern and central Asia. Bulletin of the American Museum of Natural History, 234:1-88. Wang, B., and M. R. Dawson. 1994. A primitive cricetid (Mammalia: Rodentia) from the Middle Eocene of Jiangsu Province, China. Annals of Carnegie Museum, 63:239-256. Wang, B., R. Zhai, and M. R. Dawson. 1998. Discovery of Ischyromyinae (Rodentia, Mammalia) from the Middle Eocene of North China. Vertebrata PalAsiatica, 36(1): 1-12 (in Chinese, English summary). Wood, A. E. 1962. The Early Tertiary rodents of the family Paramyidae. Transactions of the American Philosophical Society, n.s. 52(1): 1-261. . 1977. The evolution of the rodent family Ctenodactylidae. Journal of the Palaeontological Society of India, 20:120-137. ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 3, Pp. 231-232 17 August 2001 REVIEW TROGONS AND QUETZALS OF THE WORLD. Paul A. Johnsgard. 2000. Smithsonian Institution Press, Washington, D. C. 223 pp. ISBN 1-56098-388- 4 (hardcover). $49.95. Trogons and Quetzals of the World fills a significant gap in the bird literature. Trogons are a group of birds of great interest in ecology, natural history, evolution, and conservation. However, nearly one-and-a-half centuries have passed since the last publication of a monograph for this group of birds. Paul A. Johnsgard’s reputation speaks for itself. The author has written nearly 40 books, most of them covering technical aspects of particular groups of birds. He has worked closely with the Smithsonian Institution Press to produce titles such as North American Owls: Biology and Natural History, The Hummingbirds of North America, and The Pheasants of the World: Biology and Natural History. Trogons and Quetzals of the World maintains the high standards of his previous publications, surpassing the quality and information of many of them. Trogons and Quetzals of the World opens with a brief introduction in which the author gives an overview of the importance that quetzals had in the pre- Colombian cultures of Middle America, and how such visions had prevailed in modern times. In a broad sense, the book is organized into three main sections. The initial section, entitled “Comparative Biology,” covers general information of the family (20 pp.). It is divided into four topics: a) evolutionary relationships, b) anatomy and morphology, c) general behavior and ecology, and d) breeding biology and populations. Gathering all the information for this section is a com- mendable effort, but generalizations can sometimes result when dealing with a group as diverse as trogons. For instance, figures 3-5 fail in showing an apparent pattern between altitudinal and latitudinal distribution in the New World trogons. Fortunately, such flaws are unimportant and do not detract from the value of this book. The evolutionary relationships are carefully presented, and include the most recent available information for these birds. The review of the external anatomy is nicely illustrated using detailed ink drawings by the author. The second section, which represents the main body of the book (169 pp.), is focused on the species accounts. The systematic arrangement follows the classi- fication suggested by Sibley and Monroe (1990), which divides the 39 species of trogons into six genera: Apaloderma, Pharomachrus, Euptilotis, Priotelus, Tro- gon, and Harpactes. The biology of some species is well known (e.g., Skutch 1942, 1944, 1948); nonetheless, there is still much to learn from the natural his- tory and evolution of many of these birds. As Johnsgard asserts in the preface, “I am saddened by the number of times I had to enter ‘No information available’ in the species accounts [p. ix].” This section observes the classical organization of many ornithological reviews. A short commentary for every subfamily and genus is presented before the detailed species description. Every technical dossier includes a list of vernacular names; the geographic range and an updated map for the species; all the currently accepted subspecies and their distributions; a section of morphometries including measurements and weights for each subspecies; de- tailed descriptions of the adult male, female, immature, and juvenile; field and 231 232 Annals of Carnegie Museum VOL. 70 hand identification clues; a description of the geographic variation; an overview of the ecology, including habitat range, food and foraging strategies; general, social, and sexual behavior; vocalizations (no sonograms are presented); infor- mation on breeding biology, which is divided into chronology of breeding, nest sites, eggs and incubation, and brood rearing; and finally, conservation and evo- lutionary relationships. I have no doubt that ornithologists in all areas will find these summaries most valuable for their work. The third section encompasses a series of appendices (29 pp.). One presents a nice explanation for the derivations of scientific and vernacular names. A classical dichotomous key to the genera and species of trogons, taken in part from Ridgway (1911), comprises a second appendix. For those who are not well versed on technical terminology, a glossary is included. The bibliography is probably one of the most useful parts for the ornithologist. This final appendix functions as a compendium of the relevant literature written about trogons, and not merely as a list of the references cited in the text. Before Johnsgard’s book, the only available monograph of trogons was pub- lished by Gould in 1875. Gould’s work consisted of 46 lithographic plates with limited comments on the natural history of the species. Johnsgard uses 35 of Gould’s plates to illustrate one of each available species. Most of them are full- page reproductions (125 X 190 mm), and the printing quality is neatly done. For the Whitehead’s trogon, the author used a reproduction from a hand-colored lith- ograph by J. Kuelemann painted for the species original description. The remain- ing species were splendidly painted by James D. McClelland, Daniel Lane, David Riser, Dana Gardner, and John O’Neill. These illustrations alone are worth the price of the book. Recapitulating, Trogons and Quetzals of the World by Paul A. Johnsgard is clearly a book that reflects years of research, and it is one of the most complete sources of information for trogons in general. This is a technical book, but it is also a comprehensive and attractive review for both scientists and non-specialists. Although each species file includes identification guidelines, it should be empha- sized that the book itself is not an identification field guide for trogons. It is the accomplishment of a dedicated ornithologist and a book that should be included in the library of any museum or bird enthusiast. Literature Cited Gould, J. 1875. A monograph of the Trogonidae, or family of trogons. Published by the author, London, United Kingdom. Ridgway, R. 1911. The birds of North and Middle America. Part 5. Bulletin of the U.S. National Museum, 50:1-859. Sibley, C. G., and B. L. Monroe, Jr. 1990. Distribution and classification of the birds of the world. Yale University Press, New Haven, Connecticut. Skutch, a. F. 1942. Life history of the Mexican Trogon. Auk, 59:341-363. . 1944. Life history of the Quetzal. Condor, 46:213-235. . 1948. Life history of the Citreoline Trogon. Condor, 50:137-147. ALEJANDRO ESPINOSA DE LOS MONTEROS, Department of Ecologia y Comportamiento Animal, Instituto de Ecologia A. C., Km 2.5 Antigua carretera a Coatepec; Apartado Postal 63, Xalapa, Veracruz 91000, Mexico. ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 3, Pp. 233-234 17 August 2001 FROM THE ARCHIVES AND COLLECTIONS REMARKS ON THE ANNALS OF CARNEGIE MUSEUM^ S CENTENNIAL AND THE INTRODUCTION OF A NEW FEATURE W. Orr Goehring* This year marks the one hundredth anniversary of the Annals of Carnegie Museum. In a 1901 editorial introducing the journal’s first issue, then Director of Carnegie Museum William J. Holland expressed his hope that the dawn of a new century would find the Annals on library shelves around the world. With the arrival of that century, Carnegie Museum of Natural History now shares its pub- lications with over 500 libraries and institutions in more than 80 countries. During the course of those one hundred years and seventy volumes, the Annals has experienced a number of variations in both appearance and content. Each issue of the first 21 volumes, for instance, began with a section entitled “Editorial Notes,” in which Holland summarized current events at the museum, including collection efforts, new exhibits, and forthcoming publications. These notes have proven to be of great value in tracking historical aspects of the museum. In 1904, Holland also wrote the first in what would become a series of obituaries honoring individuals affiliated with the museum. “Editorial Notes” ended with Holland’s death on December 13, 1932, and the publication of obituaries became more infrequent, though one did appear in 1985. More recently, the Annals introduced a book review feature, the first example of which appeared in volume 60 in 1991. The current issue of Annals of Carnegie Museum inaugurates a new feature called “From the Archives and Collections.” Articles of this type will highlight significant material from the museum’s archives or collections, including historical texts, photographs of important specimens or artifacts contained in section col- lections, noteworthy photographs that are out of print or have never before been published, and compilations of data of historical interest. Generally, the format of each article will include a one-page description of each figure with the figure itself to appear on the following page. “From the Archives and Collections” will be published as an occasional feature, rather than as a set piece in each issue. The first two examples of this new feature appear on the subsequent pages. The first, fittingly, is a reprint of Holland’s 1901 editorial, which eloquently out- lines the mission of the scientific publications program begun with the publication of the Annals' first issue. The second discusses a previously unpublished photo- graph taken by former Curator of Archaeology and Ethnology Carl V. Hartman of an archeological excavation in Guanacaste province, Costa Rica. In the 106 years since its inception, Carnegie Museum of Natural History has amassed a wealth of collections, the study of which continues to yield important insights into our understanding of the natural world. However, the sheer volume of accumulated material precludes a published study of each specimen or artifact, Managing Editor of Scientific Publications 233 234 Annals of Carnegie Museum VOL. 70 and the passage of time dictates that some of those which have been published will be forgotten. This new feature provides a forum for introducing or, in some cases, reintroducing significant material from the past so that it can be appreciated and reevaluated in the light of the present. ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 3, P. 235 17 August 2001 FROM THE ARCHIVES AND COLLECTIONS EDITORIAL FROM ANNALS OF CARNEGIE MUSEUM, VOLUME 1, NUMBER 1 (1901) William J. Holland’ The publication of the observations and discoveries of those, who form the staff of a well-organized museum, has always been recognized as a proper func- tion of such an institution. The body of scientific literature, which has emanated from the larger museums of Europe and America, is very considerable, and com- prises some of the most important and valuable contributions, which have been made to science. Though the Carnegie Museum is the youngest of the great mu- seums of America, and may be said only quite recently to have embarked upon its career of usefulness, the labors of those connected with it have already resulted in the accumulation of a large amount of material, possessing great scientific interest and value. The study of these collections is certain to throw much light upon many questions of profound interest. For the purpose of facilitating the diffusion of knowledge gained, often laboriously and at great expense, by those who are directly connected with the museum, as well as to form a medium for the publication of the results of the labors of others, who are associated with them, or are pursuing inquiries along the same, or kindred lines, the Trustees of the Institute have authorized the publication of two series of papers, which will be designated respectively as Annals and Memoirs. The Annals will appear in octavo form, the Memoirs in quarto. They will be published from time to time as material is provided, and when pages enough to form a volume have been printed, a title-page, table of contents, and index will be supplied, and a new volume of each series will then be begun. Under the regulations, which have been adopted, twenty-five separata of articles prepared by authors will be furnished them at the expense of the institution. Should more be desired by an author, he may obtain them by giving notice, upon the acceptance of his article, of the number which he desires. The cost of printing separata over and above the number of twenty- five will be borne by the author. The present issue of the Annals is the first which comes from the press. It appears with the dawn of a new century. It is the earnest hope of all concerned in this undertaking that the birth of this journal at this auspicious time may mark the beginning of a long and successful career, and that the dawn of the next century may find in the libraries of the world a long array of volumes, which, in the wealth of information which they contain, may justify the action of those who have sanctioned the beginning of this enterprise, and may prove to be a monument to the broad philanthropy and sagacity of the noble founder of the institution — ANDREW CARNEGIE. ‘ Former Director, Carnegie Museum 235 ' V / M i» la Ti 4 •u-n KifHi-.f ’ i V' ■ ' '• ♦■■ ue.ti Ve ■. ;, ‘ u’v4;'’*i /> i; V'‘( " •.•umv’v^,n in» jAi.j pm-i ^ (lUil; ! ‘ .iK ' .1. hfA;vj^ifV'f i*mi JttuHiSm --.M y ^■ u h) Ui.'.s ^ I ff- ivt. , V, .1 . j . . . 7^? r ^ *i,rt .V'4|m, . Mj,| ifU?'t'?fUV 1»:> ^U^J tidT* 7fMv? 4)ii 7,’m* (f. • '.M' ’ ij-u M .11 V , ^ . .< ,,,-Mt>arA /aii/t ^qa/, * . , ' . ^d^■^fly^'tl ^ ru*‘w4 /K ‘ 'i'-- ' **i'i‘.f»Icr i>oi» "^/ti ■'‘^ *'^-'*“4*"' '• 4'r>onit ’*> Vjfi^ .,'.«/ ,1 , 'NMivr^.f 'Mjp i»iiv^ jrt /«fil tm« j.:>ni5«o '<■» v.’-,;'^!.:,, •it, .v' »i» ^-#1^ ‘ M:.n i t-*" »ro#irslw(5tm:> -‘ -•fH 'pP r- ■•♦n/' *,1 fill.;, ^ . ;Mf:^..llf):i •; • li yfnnv .nH y>‘/diiy fiiW 4'^ •- -^I. P'- ■'» '•* i*’ J'irr ♦*• , wqj|>-.T v>e/< f ». .'m(w iSrii-- ' >r T- . MO s ‘ iV . ir-M/ r.' ! •,!! ^fi't.'j'il *41 '.ifp •'‘I tn it J.t$ti fjTtiS ‘««i : '‘* I jtif t • Ml. ’ l.l»< ?: • . nv.V* .’u'K .‘.1 vf% 'll'. -♦!:* r»d ••nj; ’ I'dT *P7»:i»pi% rihMA^^.r' :>di '''* ''• ' i^r.i f- M-/, ^11 nfdisi ... *''’■ ", ^''' • ' •' ‘'' :* * •’' *■.-■ ?•■ f;if mrje^ 9^11 -.■^*^' ''‘'4 '‘' ■<'' ' ■ ■ p rn ^(;J^ ><-' '*>*'‘.in!t9cl br».v ' h ‘ffi ' eAmIilh.* fd iHrst4 ;f Pm: 4 iiv!tlv» »jpil ;>.Mf -p <*' .'orj lit. ii- IT ‘Mf.,4,^ ^4^ i»v/«iti ad) JjiV = MU ••;. f.Mt; /Pt,f ... ii.rrf Mit f4#if : t ti* *w1' if.i h/ia .♦•*vu;v InU ,%j ;^x;4>i k* io ^ti^DAiiPi;. 1/ ► ■ la ' ' ,r q * ^ . ■' '•*■ .. ■■•' ■‘^'M If/ / MiU m rii;if 4 »i< J!u,.!. - ti , I I f)i }.. .1,- ^lUjii J,f|» ).• jfti tiiitfurtriM Wff^ ' I /V.,- i(r.,i, ,.,!) «I, vliiiKJ 9(h H*.,t.|':::«:^m liu# .j.j-’ irftriHttrfi,} .Tu) . M - ■ 'M i ^ *’ V '! w> A I .* 1 r>,i) ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 3, Pp. 237-238 17 August 2001 FROM THE ARCHIVES AND COLLECTIONS AN EXCAVATION IN GUANACASTE PROVINCE, COSTA RICA David R. Watters ^ and Oscar Fonseca Zamora^ Carl Vilhelm Hartman served five years (1903-1908) as the first Curator of Archaeology and Ethnology at Carnegie Museum. Soon after being employed, Hartman traveled to Costa Rica on a seven-month expedition to excavate archae- ological sites and purchase collections. He dug in the Nicoya peninsula of Guan- acaste province on the Pacific side and in the Central Highlands. The only research Hartman (1907) published in detail dealt with the burial ground of Las Huacas in Nicoya. His 1907 monograph, though illustrated extensively, contains no field photographs, a point of special interest since the Section of Anthropology pos- sesses 113 images (71 glass plates and 42 “old prints”) documenting Hartman’s 1903 fieldwork, mostly Central Highlands sites. A single photograph (Fig. 1), published here for the first time, illustrates an excavation in the Pacific region. The photograph provides detailed contextual information about the artifacts. Nine metates are visible, six (#s 1-6) on the excavation floor in what appear to be pits and three (#s 7-9) removed to the ground surface. Four metates (#s 1, 7-9) are broken; five seem intact. The only other artifact (#10) is a bowl. The absence of artifacts in the profile behind the boy indicates that a layer of sterile (non-artifact bearing) overburden existed above the level containing the metates, a characteristic noted by Hartman (1907:15) for Las Huacas and other burial grounds in the Pacific region. Rounded (conical) legs, curved platforms, and limited areas (or total absence) of low-relief designs are attributes typical of the rectangular- shaped, tripod metates whose distribution is restricted to the Guanacaste region from 300 BC to AD 500. All Fig. 1 metates have curved platforms and two broken metates on the surface display three conical legs. Hartman’s Las Huacas burial ground map (1907:plate XLVII) shows 16 metates, some with carved designs, positioned singularly and in clusters. Carved designs are not evident on the Fig. 1 metates although their “in- visibility” may result simply from the photograph’s poor resolution. Is this photograph a verifiable image of Hartman’s 1903 Las Huacas excava- tion? There are reasons to support that attribution. Hartman’s writings state that Las Huacas was the only site he excavated in Guanacaste province in 1903. Fig. 1 attributes that correlate well with his excavation description include the tripod metates, their depth below ground surface, their distributions as individual and clustered artifacts, the apparent presence of pits, the sterile stratum, and the tem- poral concordance between the span of metate manufacture and use of the cem- etery. Yet, there are discrepancies. Placement of metates in Fig. 1 differs from their depiction on his map. No individual metate in the photograph can be cor- related, for certain, with a metate in the Las Huacas collection (the photograph’s lack of visible design elements frustrates any match). Absent in the photograph are other kinds of artifacts and the skeletons seen on the map. ’ Curator, Section of Anthropology ^ Research Associate, Section of Anthropology (and Professor, University of Costa Rica) 237 238 Annals of Carnegie Museum VOL. 70 Fig. 1. — Modern print (cropped) from glass-plate negative (G1029) in the Section of Anthropology. The authors conclude that this image unquestionably depicts an excavation in Guanacaste province but, because of the inconsistencies, it cannot be verified to be Las Huacas. If the image is Las Huacas (a strong possibility), the site sector differs from that shown on Hartman’s map. Whatever site it depicts, Fig. 1 is important historically, as a rare image photographically documenting in situ ar- tifacts from the earliest days of archaeological research in Costa Rica. Acknowledgments: Research supported by the Adrienne and Milton Porter Charitable Foundation. Literature Cited Hartman, C. V. 1907. Archeological Researches on the Pacific Coast of Costa Rica. Memoirs of the Carnegie Museum, 3(1). INSTRUCTIONS FOR AUTHORS ANNALS OF CARNEGIE MUSEUM consists of contributions to the earth sciences (including paleontol- ogy), life sciences, and anthropology, in 30 by 46 pica format (127 by 195 mm or 5 by 7% inches). Submit all manuscripts to the Office of Scientific Publications. Au- thors should give particular attention to scientific con- tent, format, and general style for the ANNALS. Manu- scripts that do not conform to the style of the AN- NALS will be returned to the author immediately. Every manuscript will be subjected to external peer re- view. Authors should submit the names, addresses, phone and fax numbers, and e-mail addresses of at least four qualified potential reviewers for each manuscript. Authors will be asked to subsidize, if fimds are available, any or all costs of publication (approximately $ 1 00/page printed). A detailed set of instructions to authors may be requested from the Office of Scientific Publications. Manuscript Style. — ^Articles should include the fol- lowing items in this order: title page, abstract, text (with desired headings), acknowledgments, literature cited, ta- bles, figure captions, and copies of illustrations. All manuscripts must be typed double-spaced on standard 814- by 1 1-inch white bond paper, with at least one-inch margins all around, and submitted in quadruplicate — an original , for the editors, and three review copies. All pages should be numbered, including tables, literature cited, and the list of figure captions. Only correspon- dence quality or better dot matrix printouts will be ac- cepted; draft mode is unacceptable. A monospaced font, (e.g.. Courier) is required. A disk in WordPerfect x.x or ASCII is helpful. Title Page. The title should be brief, include the ani- mal or plant group involved, be placed on a separate page, and appear two to three inches below the top mar- gin. Include the author’s name(s) and the affiliations of non-CMNH authors. In the case of multiple authorship, indicate the address to which proofs should be sent. Abstracts. Abstracts should be short, but substantive, and included at the head of the first page of text. Text. Do not right justify text or break (hyphenate) a word at the end of a line. Footnotes and acknowledg- ments as footnotes are unacceptable. All text is double- spaced in a monospaced font (e.g.. Courier). Tables and Figure Legends. The list of figure legends and each table should be typed consecutively on indi- vidual pages separate from the text. Tables must be dou- ble-spaced throughout with no vertical lines. Measurements. Metric units should be used, if possi- ble. Literature Cited. All references to literature in text must appear in the Literature Cited section. The data (author, date, and page) in both citations must agree. Do not abbreviate the titles of periodicals or serials. The following style, double-spaced, should be used in Lit- erature Cited: 1) Two authors in a journal series: Soltis, D. E., and P. S. Soltis. 1992. The distri- bution of selfing rates in homosporous ferns. American Journal of Botany, 79:97—100. 2) Same authors repeated — ^use three-em dash: . 1923. The fauna of the Ardyn Obo For- mation. American Museum of Natural History Novitates, 98:1-5. 3) Same authors plus a third author — repeat all authors: Knutson, L. V, R. E. Orth, and W. L. Murphy. 1986. Catalog of Sciomyzidae (Diptera) of America north of Mexico. Entomography, 4: 1-53. 4) Chapter in an edited volume: Rausch, R. L. 1963. A review of the distribution of Holarctic mammals. Pp. 29-43, in Pacific Basin Biogeography (J. L. Gressitt, ed.). Bishop Museum Press, Honolulu, Hawaii. 5) Unpublished dissertation: Smith, J. P. 1976. Review of Eocene Mammals. Unpublished Ph.D. Dissert., University of California, Berkeley, California. 6) Book: White, M. J. D. 1961. The Chromosomes. Meth- euen and Co., Ltd., London, United Kingdom. 7) Journal articles with usual volume and issue number: Anderson, W. I. 1969. Lower Mississippian con- odonts from northern Iowa. Journal of Pale- ontology, 43:916-928. Figures. Including all illustrative materials (line art, halftones, photographs), figures are to be numbered in Arabic numerals. Four sets of figures are required, one (original artwork) for reproduction, three for reviewers. Photocopies of photographs for reviewers are usually not acceptable but are adequate for line drawing review cop- ies. Figures may not be larger than 17 by 12 inches. Reducing figures is the responsibility of the author. All figures must be reducible to a maximum of 127 by 195 mm (30 by 46 picas) without loss of clarity. Line copy should be designed for reduction to % or Vi or actual size. Typewritten figure copy will not be accepted. Pho- tographic figures should be submitted at actual repro- duction size, if possible. Rectangular halftone figures should be abutted, with- out intervening spaces. The printer will insert narrow white spaces during the reproduction process. All figures must have minimally one-inch borders all around. Each figure should be given a protective cover and identified on the back side. Lettering and/or a magnification scale (linear metric scale) for rectangular halftone figures should be placed directly on the photo, not in a blank space between pho- tos. The scale or lettering for closely cropped photos can be placed in blank areas close to the figure. Proof. — The author should answer all queried proof marks and check the entire proof copy. Return corrected page proof with the edited manuscript promptly to the Office of Scientific Publications. If an author chooses to make extensive alterations to a paper in proof stage, the author will bear the cost. Original manuscripts will not be returned unless request- ed. Figures will be returned to the author only if re- quested prior to publication. ISSN 0097-4463 AQ 36p ?-7K 'Annals of CARNEGIE MUSEUM THE CARNEGIE MUSEUM OF NATURAL HISTORY 4400 FORBES AVENUE • PITTSBURGH, PENNSYLVANIA 15213 VOLUME 70 24 MAY 2001 NUMBER 2 CONTENTS ARTICLES The skull of Mesenosaurus romeri, a small varanopseid (Synapsida: Eupelycosauria) from the Upper Permian of the Mezen River Basin, northern Russia Robert R. Reisz and David S Berman 113 A diadectid (Tetrapoda: Diadectomorpha) from the Lower Permian fissure fills of the Dolese quarry, near Richards Spur, Oklahoma Robert R. Reisz and Tammy E. Sutherland 133 Rhizomyidae from the Lower Manchar Formation (Miocene, Pakistan) .... Wilma Wessels and Hans de Bruijn 143 Fleas (Siphonaptera: Ctenophthalmidae and Rhopalopsyllidae) from Argentina and Chile with two new species from the Rock Rat, Aconaemys fuscus, in Chile Michael W. Hastriter 169 Editors, ANNALS, BULLETIN and SPECIAL PUBLICATIONS: D. S. Berman, Ph.D. D. R. Watters, Ph.D. J. R. Wible, Ph.D. W. O. Goehring, Managing Editor Manuscripts, subscriptions, orders for individual numbers, and changes of address should be sent to; Office of Scientific Publications Carnegie Museum of Natural History 4400 Forbes Avenue Pittsburgh, PA 15213-4080 Phone (412) 622-3287 Fax (412) 622-8837 E-mail: scipubs@clpgh.org ANNALS OF CARNEGIE MUSEUM is published quarterly by The Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania 15213-4080, by the authority of the Board of Trustees of Carnegie Institute. © 2001 Carnegie Institute. @ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 2, Pp. 113-132 24 May 2001 THE SKULL OF MESENOSAURUS ROMERI, A SMALL VARANOPSEID (SYNAPSIDA: EUPELYCOSAURIA) FROM THE UPPER PERMIAN OF THE MEZEN RIVER BASIN, NORTHERN RUSSIA Robert R. Reisz^ Research Associate, Section of Vertebrate Fossils David S Berman Curator, Section of Vertebrate Paleontology Abstract Restudy of Mesenosaurus romeri, based on new and previously described cranial materials from the Upper Permian of the Mezen River basin of northern Russia, confirms its assignment to the synapsid eupelycosaurian family Varanopseidae. Comparisons with other members of the family support a pattern of relationship that recognizes two clades: one is composed of Mesenosaurus and Mycterosaurus for which the subfamily designation Myctersaurinae is proposed, and the other includes the remaining well- known varanopseids Elliotsmithia, Varanops, Varanodon, and Aerosaurus for which the subfamily des- ignation Varanodontinae is proposed. Among the late Paleozoic synapsids, Varanopseidae has the longest fossil record, extending from the end of the Carboniferous to well into the Late Permian, and the widest geographical distribution, including North America, South Africa, and Russia. Key Words: Varanopseidae {Mesenosaurus), Synapsida, Upper Permian, Mezen River Basin, Russia Introduction The Mezen River basin of northern Russia has extensive exposures of Upper Permian sediments along the edges of several rivers, especially the Peza and Kimja rivers, both affluents of the Mezen River. These sediments, although visited only sporadically by paleontologists and geologists, have produced the skeletal remains of a diverse assemblage of amniotes, including numerous enigmatic par- areptiles, at least two therapsids, and most interestingly a small synapsid of var- anopseid affinities, Mesenosaurus romeri Efremov (1938). Mesenosaurus romeri was originally described on the basis of a partial skull, but its assignment by Romer and Price (1940) to the synapsid family Varanopseidae was only tentative, owing to the incompleteness of the holotype and only known specimen. A few additional specimens of M. romeri were recovered in the 1950s, which led to its more recent restudy by Ivachnenko (1978). These new specimens, including a poorly preserved articulated skeleton, were the basis of Ivachnenko ’s argument that Mesenosaurus was the oldest known archosaur (Evans, 1988; Carroll, 1988). As part of a systematic program of collecting in this area, several new specimens have been recovered and prepared, allowing a reevaluation of the anatomy and phylogenetic relationships of this interesting Paleozoic amniote. This study reaf- firms the assignment of Mesenosaurus to the synapsid family Varanopseidae. Anatomical structures are identified by the following abbreviations: an, angular; bo, basioccipital; co, posterior coronoid; d, dentary; ec, ectopterygoid; ex, exoc- ' Department of Biology, Erindale Campus, University of Toronto, Mississauga, Ontario L5L 1C6, Canada. Submitted 15 December 1999. 113 14 Annals of Carnegie Museum VOL. 70 cipital; f, frontal; j, jugal; 1, lacrimal; m, maxilla; n, nasal; o, opisthotic; p, parietal; pal, palatine; pf, postfrontal; pm, premaxilla; po, postorbital; pp, postparietal; pra, prearticular; prf, prefrontal; ps, parasphenoid; q, quadrate; qj, quadratojugal; sc, sclerotic element; so, supraoccipital; sp, splenial; sq, squamosal; st, supratemporal; su, surangular; t, tabular; v, vomer. Systematic Paleontology Class Amniota Subclass Synapsida Eupelycosauria Kemp, 1982 Family Varanopseidae Romer and Price, 1940 Mycterosaurinae, new subfamily Definition. — Varanopseid synapsids more closely related to Mycterosaurus than to Varanops. Diagnosis. — Small varanopseid synapsids characterized by greatly expanded dor- sal lamina of maxilla that contacts the prefrontal, resulting in the anterior shortening of the lacrimal to a level less than half the distance from the orbit to the naris and the loss of a nasal-lacrimal contact; prefrontal with well-developed ventral orbital process that contacts the palatine; paroccipital process of opisthotic anteroposterally, rather than dorsoventrally, expanded oval in cross section; caniniform region located far forward and at a level immediately behind the external naris. Mesenosaurus Efremov, 1938 Type species Mesenosaurus romeri Efremov, 1938 Revised Diagnosis. — Myterosaurine eupelycosaur characterized by the follow- ing cranial features: 1) premaxilla slender and with mate forms a narrowly rect- angular snout in dorsal and ventral views; 2) dorsal process of premaxilla long and forms anterior half of dorsal margin of external naris; 3) deep excavation of the lateral surface of the body of the premaxilla narrows the base of the dorsal process to produce an expanded narial shelf that extends nearly to the snout tip; 4) palatal process of premaxilla with unusually long median suture; 5) well-de- veloped depression on the lateral surface of the nasal that extends posteriorly from the narial border to nearly the anterior end of the prefrontal; 6) slight lateral swelling of the maxilla at the level of the caniniform tooth; 7) short posterior process of the maxilla fails to reach the level of the postorbital bar; 8) first pre- maxillary tooth smaller than the second and third teeth; 9) single, median vo- merine tooth row; 10) postorbital cheek region of skull unusually broad and low, with nearly vertical posterior margin; 11) posterior edge of transverse flange of the ptyergoid is angled slightly anterolaterally from basal articulation; 12) stapes slender, short, and rodlike, with modestly developed footplate and distally ex- panded quadrate process. Mesenosaurus romeri Efremov, 1938 Holotype. — PIN (Paleontological Institute, Russian Academy of Sciences, Mos- cow) 158/1, partial skull and nearly complete right mandible (Fig. 1). Referred Specimens. — PIN 3586/8a, partial skull of large individual (Fig. 2); PIN 3706/11, 3706/ 15, partial skulls of juvenile individuals (Fig. 3, 4), SGU (Saratov Geological Institute, Russia) 104 V/1558, partial skull, similar in size to the holotype (Fig. 5). 2001 Reisz and Berman — Late Permian Varanopseid from Russia 1 Fig. 1. — Mesenosaurus romeri, holotype PIN 158/1. A. Skull in dorsal view. B. Right mandible medial view. C. Right mandibe in lateral view. Scale = 1 cm. 116 Annals of Carnegie Museum VOL. 70 Fig. 2. — Mesenosaurus romeri, PIN 3586/8a, anterior part of largest known skull in ventral and dorsal views. Scale = 1 cm. Horizon and Locality. — Mezen River Basin, northern Russia, Lower Tatarian, Upper Permian. Diagnosis. — Same as for genus. Description and Comparisons Skull. — General. The reconstruction of the skull shown in Figure 6 is a com- posite based primarily on the holotype, relying on the referred specimens only when necessary. Details of the snout region were available only in the larger specimens, whereas the posterior portions of the palate and braincase were well preserved only in the smaller, juvenile specimens. The pattern of the dentition was based largely on more mature specimens. As discussed below in the Discus- sion section, the Varanopseidae is recognized as being divisible into two subfam- ilies: the stem-based Mycterosaurinae is proposed for Mycterosaurus and Mes- enosaurus, whereas all other known genera, Elliotsmithia, Aerosaurus, Varanops, and Varanodon, are included in the proposed stem-based Varanodontinae. With reference to this subdivision, the following description not only compares Mes- enosaurus with Mycterosaurus, but also emphasizes features defining Varanodon- tinae as including taxa which are more closely related to Varanodon than to Mycterosaurus. 2001 Reisz and Berman — Late Permian Varanopseid from Russia 117 Fig. 3. — Mesenosaurus romeri, PIN 3706/11, nearly complete skull of juvenile in dorsal, and ventral views. Scale = 1cm. In profile the skull has a low, subrectangular outline, with the occipital margin being normal to the jaw line. The skull outline is very distinctive in dorsal view, with the cheek regions diverging widely to about the level of the postorbital bar, where a sharply defined angulation then orients the temporal margins parallel to the midline. The snout of Mesenosaurus is unique among Paleozoic amniotes in being formed essentially by only the premaxilla and having a narrowly rectangular outline with a truncated tip in dorsal view. The orbit is unusually large and appears anteroposteriorly elongate because of the reduced height of the skull. The dorsal rim of the orbit is expanded slightly above the skull table as a rounded ridge. The parapineal foramen is large and located close to the posterior border of the skull roof. The lateral temporal fenestra is tall, occupying nearly the entire height of the skull. A pronounced sculpturing, consisting of a distinct pattern of grooves, covers most of the skull roof. In addition, a welLdeveloped tubercular or nodu- larlike ornamentation extends along the orbital margins of the prefrontal, post- orbital, and jugal. The internal nares are greatly elongated, equaling one-half the length between the snout tip and the anterior margin of the subtemporal fossa, and the palatal surface bears a complex pattern of tooth-bearing ridges. The un- usually slender proportions of the lower jaw match those in other varanopseids. Skull Roof. — The premaxilla, present in the holotype (Fig. 1) and PIN 4586/8 (Fig. 2), is large and possesses a minimum of five marginal teeth. In both dorsal or ventral views the paired premaxillae form a narrow, abruptly truncated snout tip with nearly parallel lateral margins, giving it a rectangular outline. Just above 118 Annals of Carnegie Museum VOL. 70 Fig. 4. — Mesenosaurus romeri, PIN 3706/15, partial skull in dorsal and occipital views. Scale = 1cm. the marginal tooth row the lateral surface of the body of the premaxilla is deeply excavated so as to produce an expanded narial shelf that extends nearly to the tip of the snout. The excavation results in a narrowing of the base of the dorsal process, which is otherwise well developed in both length and width. The pro- cesses contact one another throughout their length, expanding slightly as they form the anterior half of the dorsal margin of the greatly elongated external nares, 2001 Reisz and Berman — Late Permian Varanopseid from Russia 19 Fig. 5. — Mesenosaurus romeri, SGU 104V/1558, nearly complete skull with right mandible in dorsal and lateral views. Scale — 1 cm. then narrowing as they extend between the nasals to a level well beyond the posterior borders of the external nares. The premaxilla forms nearly the entire ventral margin of the external naris. In contrast to all other late Paleozoic amni- otes, there is no sharp, angular union between the dorsal and lateral surfaces of its subnarial bar, but rather, as seen only in varanopseids, the external surface of the subnarial bar is broadly rounded in transverse section. As on the skull roof, the extraordinarily narrow snout appears to be responsible for the extensive palatal contact between the premaxillae. Here their midline union extends over half the anterior, palatal length of the bone. Their remaining, posterior portions form a short palatal process which are narrowly separated along the midline by anterior 120 Annals of Carnegie Museum VOL. 70 2001 Reisz and Berman — Late Permian Varanopseid from Russia 121 processes of the vomers before extending onto the ventral surface of the vomers. The premaxillary contribution to the internal naris is restricted to a very small portion of the anterior lateral border. A maximum of five teeth is borne by the premaxilla: the first is smaller than the second and third, but larger than the last two. The teeth are similar to those in other varanopseids in being closely spaced, strongly recurved, sharply pointed, and flattened from side to side with the larger teeth possessing a well-developed cutting edge along the distal half of the pos- terior edge. The nasal exhibits a unique, well-developed depression that extends posteriorly from its anterolateral margin bordering the external naris to nearly the anterior end of the prefrontal (Fig. 2B). The width of the nasal in this region is narrowed greatly by the dorsal process of the premaxilla. A wide abutment, rather than the typical overlapping suture, marks the contact between the nasal and the dorsal lamina of the maxilla. The lateral margin of the broader, posterior half of the nasal is gently bowed ventrally to its contact with the maxilla and prefrontal. The frontal, preserved in all the specimens, exceeds slightly the nasal as the longest bone of the midline series, and its anterior process exceeds greatly the posterior process in width. As in other varanopseids, the contribution of the frontal to the orbital margin is extensive and is achieved by a medial emargination at the orbits, rather than by a lateral extension or lappet of the frontal as in sphenacodontids. The posterior process of the frontal is like that in Mycterosaurus in forming a narrow, triangular extension that diverges from the midline as it contacts the medial margin of the postfrontal. The broad parietals not only occupy most of the postorbital skull table (Fig. 3, 5), but also form a broadly triangular, anterior, midline process that extends well into the supraorbital region. The occipital mar- gin of the parietal is broadly concave, with the posterolateral corner being drawn out into a winglike process. A well-developed occipital flange of the parietal is overlapped externally by the postparietal and tabular and therefore is not visible in the articulated skull. A deep, narrow groove on the posterolateral wing of the parietal received the anterior portion of the supratemporal; a distal portion of the supratemporal, not represented in any of the specimens, is presumed to have overlapped the squamosal, A short, narrow strip of the parietal posterolateral wing is exposed dorsally between the supratemporal and postorbital and is bordered distally by the squamosal. The unusually large, transversely oval parapineal fo- ramen lies close to the posterior margin of the skull table. The large, roughly rectangular postparietals are restricted entirely to the occiput, have a deeply con- cave occipital surface, and unite in a median occipital ridge (Fig. 5). The median ridge ends just short of the ventral margin of the postparietals, below which each possesses a small, distinct, ventral medial process. Reexamination of the holotypic skull of Mycterosaurus reveals clearly the presence of paired postparietals and the same small process that defines the ventral limit of the median ridge. In lateral view the ventral margin of the premaxilla and anterior half of the maxilla describe a straight, horizontal line, whereas more posteriorly there is a very slight dorsalward angulation (Fig. 5). The maxilla is long, extending to nearly the level of the postorbital, and widely separated from the quadratojugal by the Fig. 6. — Reconstruction of Mesenosaurus romeri. Skull in dorsal, lateral with mandible, and palatal views. Teeth shown as basal cross section. Scale = 1 cm. 122 Annals of Carnegie Museum VOL. 70 jugal on the ventral margin of the skull. This is strongly contrasted by the pattern seen in all other known varanopseids, including Mycterosaurus, where the maxilla extends nearly to the level of the midlength of the subtemporal bar to contact the quadratojugal and exclude the jugal from the ventral margin of the skull. As in Mycterosaurus, however, the dorsal lamina of the maxilla is greatly expanded above the caniniform region to occupy a broadly rectangular area that not only excludes the lacrimal from the naris, but also shortens its length to a little over 40% of the distance between the orbit and the naris (Fig. 2, 5). In addition, the dorsal lamina is similar to that of Mycterosaurus and early therapsids in having a sufficient expansion to contact the prefrontal and prevent the lacrimal from contacting the nasal. The marginal dentition of the maxilla in the holotype PIN 158/1 includes 23 teeth and spaces. However, the posteriormost portion of the maxilla is incomplete, so the maximum number may have been slightly greater. The first tooth is nearly the length of the last premaxillary tooth but much smaller in basal diameter; the succeeding three teeth increase dramatically in size to a dominant caniniform tooth, and the postcaniniform teeth decrease steadily in size posteriorly. The general tooth morphology is identical to that of the premaxillary teeth. However, the well-preserved canines, as well as several postcanine teeth in PIN 3586/8a, exhibit very delicate serrations along the anterior and posterior cutting edges. It is likely that these serrations were present on all the teeth, but were lost during preparation. Confirming this, the mechanically prepared teeth of the right maxilla show no evidence of serrations, whereas those of the much more fragmentary left maxilla, which were exposed using nonmechanical methods, ex- hibit serrations. The serrations are so delicate and fine as to be not easily recog- nizable in other varanopseids. Contrary to the condition in Mesenosaurus, in all other varanopseids in which the maxillary dentition is known a caniniform region rather than a single, dominant caniniform tooth is exhibited. The state of this feature in Elliotsmithia is unknown due to the incompleteness of the holotype and only known skull (Dilkes and Reisz, 1996), The greatly reduced, subrectangular lacrimal makes only a narrow contribution to the anteroventral corner of the orbit (Fig. 5). As in Mycterosaurus, the lacrimal duct opens on the lateral surface of the skull near the orbital margin. Although poorly developed and not visible in lateral view, the suborbital process of the lacrimal contacts the jugal along the medial surface of the maxillary orbital margin. The prefrontal, well preserved in all the specimens, is a large element with a broad, well-developed ventral, orbital process that nearly excludes the lacrimal from the orbit, then continues across the medial surface of the lacrimal to contact the dorsal surface of the palatine. As in other varanopseids, the prefrontal is divided longitudinally by an abrupt right-angled bend into distinct dorsal and lateral components. In Mesenosaurus the union between the two surfaces is ac- centuated by a strongly developed, tubercular or nodularlike ornamentation just anterior to the orbit. As a result, this area of the prefrontal extends outward to overhang slightly the lateral surface of the skull. Unfortunately, this area is not well preserved in Mycterosaurus. The small, subtriangular postfrontal is like that in Mycterosaurus in being restricted almost entirely to the dorsal skull table. The postorbital is similar to the prefrontal in being divided into distinct dorsal and lateral components (Fig. 1, 3, 5). This division is also accentuated by prominent tubercular or nodularlike ornamentation at the posterodorsal corner of the orbit, as well as by a depression on the dorsal surface. Although damaged, the postor- bital in Mycterosaurus also exhibits all of the above features. In Mesenosaurus, 2001 Reisz and Berman — Late Permian Varanopseid from Russia 123 however, the ornamentation is so pronounced that the orbital rim of the postfrontal protrudes into the orbit. Despite the posterior process of the postorbital being long and broad, it is separated from the supratemporal by the parietal and squamosal. As in all other varanopseids, the jugal is triradiate with slender, splintlike pro- cesses that form the ventral borders of the orbit and temporal fenestra, and the ventral half of the postorbital bar (Fig. 3, 4). Its anterior contact with the lacrimal is narrow and, as it occurs on the medial surface of the maxilla, is visible only in dorsal aspect of the skull. In strong contrast to the condition in other varan- opseids, including Mycterosaurus, the jugal contributes to the ventral margin of the skull. This appears to be the result of the maxilla in Mesenosaurus not having the extraordinarily long posterior extension so characteristic of other varanopseids, and therefore, failing to contact the quadratojugal. A distinctive feature of the jugal is the presence of tubercular or nodularlike ornamentation at the postero- ventral corner of the orbit. It is likely that similar ornamentation was also present in Mycterosaurus, but damage and/or poor preservation prevents confirmation. The quadratojugal is a small element that contacts the jugal anteriorly and is overlapped slightly along its dorsal margin by the squamosal (Fig. 4). Posteriorly the quadratojugal exhibits an abrupt constriction along its dorsal margin as it wraps medially a short distance across the posterior surface of the squamosal. The quadratojugal then expands again to approximately its original height as it continues a short distance farther onto the posterior surface of the quadrate to interpose between the ventral margin of the quadrate foramen and the dorsal margin of the quadrate condyle. As in Mycterosaurus and Elliotsmithia, but in strong contrast to Varanops, Aerosaurus, and Varanodon, the quadratojugal is excluded from the ventral margin of the lateral temporal fenestra by an angular, antero ventral extension of the squamosal. The squamosal is roughly rectangular, but with its anterior margin deeply emarginated by the temporal fenestra (Fig. 1), creating anterior extensions of the squamosal above and below the fenestra that contact the postorbital and jugal, respectively. At the posterodorsal comer of the temporal region the squamosal curves slightly medially to make a narrow contri- bution to the dorsolateral corner of the occipital surface of the skull. Otherwise, the greater portion of the posterior margin of the squamosal is restricted to the lateral surface of the skull, rather than wrapping medially onto the posterior sur- face of the quadrate to form the occipital flange typically seen in early synapsids. This allows the posterior margin of the well-developed, bladelike dorsal process of the quadrate in Mesenosaurus to be visible in occipital view (Fig. 4). The dorsal process of the quadrate curves anteromedially as it contributes to the pos- terior, medial wall of the adductor chamber. Although most of the anterior margin of the dorsal process of the quadrate is not visible in any of the specimens studied, a conspicuous thickening defines its entire ventral margin, as it extends from the medial side of the condyle to contact the lateral surface of the posterior margin of the quadrate ramus of the pterygoid. A prominent quadratojugal foramen is located at the sutural intersection of the quadratojugal, squamosal, and quadrate. The condylar surface of the quadrate is not fully exposed in any of the specimens, but appears to be divided unequally into a smaller lateral and a larger medial condyle. The supratemporal is rarely preserved in Mesenosaurus, and then is represented only by its anterior portion (Fig. 1, 3). As in other eupelycosaurs, this element appears to be a slender strip of bone whose anterior portion is seated in a deep, narrow groove on the parietal. It was the empty anterior, parietal groove that was 124 Annals of Carnegie Museum VOL. 70 misinterpreted as an upper temporal fenestra by Ivachnenko (1978). What appears to be a posterior portion of the groove for the supratemporal is faintly visible on the squamosal of PIN 3706/1 1 (Fig. 3). The tabular, preserved only in two spec- imens (Fig. 4, 5), is a narrow, somewhat sickle-shaped element whose convex anterodorsal margin conforms to the curvature of the occipital margin of the parietal with which it makes contact. Posteroventrally, however, the tabular ends in a small, distinct, posteromedially directed, hooklike process which appears to be present also in advanced varanopseids (Reisz, personal observation). The tab- ular is unknown in Mycterosaurus. Palate. — Little is known about the palate of varanopseids, thus greatly restrict- ing or even eliminating comparisons between that of Mesenosaurus (Fig. 6C) and other members of the family. The long, narrow internal nares are separated from one another by slender, elongate vomers (Fig. 2). For a short distance posterior to its extensive contact with the premaxilla, the vomer narrows slightly, then gradually expands to its sharply notched posterior end for the reception of the anterior end of the palatine. Beginning a short distance posterior to the premaxilla, the ventral surface of the vomer forms a deep, troughlike channel which parallels the midline closely. Possibly because of the narrowness of the snout, only one of the paired vomers bears a row of teeth on a prominent ridge that extends along most of their midline contact. At the point where the vomers diverge posteriorly from the midline, however, a second tooth row branches from the main, median row, and both angle slightly laterally for a short distance along the remaining medial margin of the vomers. Both rows of teeth are then continued directly posteriorly on the pterygoids. Although in the largest specimen, PIN 3596/8A, the vomer is paired throughout its length (Fig. 2), the condition in the partial holotypic skull PIN 158/1 (not figured) appears to be different. Here, only the anterior portions of the vomers are preserved, and they exhibit a midline suture separating only the paired, slender, vermiform anterior processes wedged between the posterior palatal processes of the premaxillae; the more posterior, internarial portions of the vomers are not suturally divided. The absence of the posterior, midline suture is especially evident in the short, edentulous region between the paired, anterior processes and the single, median tooth row, where only a weakly developed ridge is present. The palatine is a large element with extensive contacts with the pterygoid medially and the maxilla laterally (Fig. 2, 3). A deep, narrow notch on its anterior lateral margin forms the posterior corner of the internal naris, whereas its convex posterior margin incises deeply into the ectopterygoid. In contrast to most other synapsids, the palatine in Mesenosaurus is not a simple, sheetlike bone, but rather is distinguished by three prominent ventral ridges. Two of the ridges extend along its lateral and medial margins, with the former being much narrower and bordering the contact with the maxilla, whereas the latter is much wider and supports a field of small teeth. These two ridges are connected by a short, narrow, transverse ridge whose medial end also carries a few teeth. The ventral surface of the palatine is deeply concave between the ridges, partic- ularly adjacent to the transverse ridge. Two foramina pierce the ventral surface of the palatine, one in the posteromedial angle formed by the lateral and transverse ridges and the other a short distance anterolaterally and adjacent to the palatine- maxillary suture. In ventral view the ectopterygoid appears as a relatively simple, sheetlike bone that occupies a small, somewhat hourglass-shaped area. From a rather broad, lateral contact with the maxilla the ectopterygoid narrows to just beyond its mid- 2001 Reisz and Berman — Late Permian Varanopseid from Russia 125 length as it extends posteromedially, then expands again slightly before ending in a widely angular incisement in the anterolateral margin of the pterygoid just anterior to the transverse flange. The free, posterior edge of the ectopterygoid, forming the concave, anterior extension of the subtemporal fossa, has a smoothly rounded, ventral ridge. There are no teeth on the ectopterygoid. In dorsal view the ectopterygoid has a narrow anterior process, visible in PIN 3586/8 (Fig. 2), that not only covers the posterior half of the lateral margin of the palatine, but also contacts the medial surface of the jugal. The pterygoid is the largest bone of the palate, forming most of the palatal surface (Fig. 2, 3). Three ridges bearing small teeth radiate across its ventral surface from a point near the basipterygoid articulation. A single row extends immediately adjacent to the medial edge of the pterygoid and is smoothly con- tinuous with the medial tooth row of the vomer. A second tooth row extends anterolaterally, beginning as a single row before expanding slightly into a narrow field which is smoothly continuous with that extending along the medial margin of the palatine. The posterior edge of the transverse flange supports a third row of about eight teeth. They are much larger than any of the other palatal teeth, although the last two or three teeth at either end of the series may decrease considerably in size. In addition, a series of smaller teeth is typically present immediately anterior to the lateral half of the principal tooth row. The palatal surface of the pterygoid between its toothed ridges is slightly concave. The lateral margins of the anterior or palatal rami of the pterygoids converge strongly toward the midline, where their sharply pointed ends wedge between the vomers. A short, narrow interpterygoid vacuity is partially closed posteriorly on either side of the midline by a stout, posteromedially projecting, triangular basal process of the pterygoid. The transverse flange occupies the same plane as the palate proper and therefore does not project below the level of the ventral rim of the skull. The posterior edge of the transverse flange angles slightly anterolaterally from the level of the basipterygoid articulation. In palatal view the proximal portion of the quadrate ramus of the pterygoid extends posteriorly, whereas distally it curves slightly laterally. Its thickened ventral edge is smoothly rounded. It is not possible to make any useful comparisons with the palate of Mycterosaurus, as only the dorsal surface of its pterygoid is preserved, and then only as an impression (Ber- man and Reisz, 1982). Braincase. — The fused parasphenoid and basisphenoid, the basiparasphenoid complex, is well preserved in PIN 3706/11 (Fig. 3), but only the parasphenoid component, which covers most of the ventral surface of the braincase, is visible. The narrow, long cultriform process is V-shaped in cross section. Its proximal portion, which divides the interpterygoid vacuity, supports a single row of small teeth, indicating that it occupied a level closely approximating the palatal plane. The precise extent of the distal edentulous portion, which would have extended dorsal to the pterygoids, is unknown. The laterally projecting basipterygoid pro- cesses have the form of stout, subtriangular plugs. An articular facet on the an- terior surface abuts against the posterior surface of the basal process of the pter- ygoid. As in other varanopseids, at the level of the basipterygoid process the parasphenoidal plate expands abruptly posteriorly to about two-thirds its maxi- mum width, then gradually expands to its posterior extent. In a manner identical to that in Mycterosaurus, the lateral margins of the parasphenoidal plate form prominent, ventrally expanded, rounded ridges which broaden slightly as they extend posteriorly to form the basisphenoidal tubera. In both forms the tubera 126 Annals of Carnegie Museum VOL. 70 merge smoothly with the ventral keel of the cultriform process and each supports proximally a row of teeth like those on the keel. Between the tubera the paras- phenoidal plate is deeply excavated, with the excavation deepening slightly on either side of a low, narrow, median ridge. The parasphenoid ends posteriorly in a feathered edge that ventrally overlaps the basioccipital. A complete basioccipital-exoccipital complex with a well-preserved occipital condyle is present only in PIN 3706/15 (Fig. 4). As in other eupelycosaurs, the condyle is roughly heart shaped, with the exoccipitals forming the dorsolateral corners of the condyle. They therefore also form the ventral and lateral margins of the foramen magnum, whereas this opening is bounded by the supraoccipital dorsally. As in Mycterosaurus, the supraoccipital is basically broad and flat, with only a slight median ridge. In contrast to the condition seen in larger varanopseids (Langston and Reisz, 1981), however, the dorsolateral process of the supraoccip- ital, usually extending above the posttemporal fenestra, is poorly developed. Al- though this area is incompletely preserved in Mycterosaurus, enough of the su- praoccipital exists to indicate a similar configuration to that in Mesenosaurus. As in other synapsids, the opisthotic has a well-developed ventromedial process that not only contacts the lateral surface of the exoccipital, but also abuts ventrally against the basioccipital-parasphenoid complex. The morphology of the paroccip- ital process of the opisthotic also appears to be similar in Mycterosaurus and Mesenosaurus. In strong contrast to the condition seen in the larger varanopseids (Langston and Reisz, 1981), the paroccipital process is not dorsoventrally ex- panded and bladelike, but rather is nearly rodlike, with a slightly anteroposteriorly expanded oval outline in cross section. The stapes, preserved in place in PIN 3706/11 (Fig. 3), exhibits a morphology that is strikingly different from those of other Permo-Carboniferous synapsids. It is a relatively slender, short, rodlike element except for a modestly developed, proximal footplate and distally expanded quadrate process. Adjacent to the foot- plate a large stapedial foramen pierces the shaft anteroposteriorly. A small rudi- ment of the dorsal process projects from the base of the footplate. The stapes resembles more closely those in early therapsids contemporaneous with Mesen- osaurus, than those in the Permo-Carboniferous eupelycosaurs. For example, the stapes in Mycterosaurus retains a large, massive footplate, a narrow neck between the footplate and a massive shaft, and a large dorsal process distal to the neck (Berman and Reisz, 1982). Mandible. — The lower jaw is very elongate and slender, matching the propor- tions of the skull (Fig. 1, 6). In lateral view the tooth-bearing margin is only very slightly concave, and the coronoid eminence is very low, rising only slightly above the dentition. The ventral margin is nearly straight and only very slightly convex. The dentary, exposed principally in lateral view, occupies 80% of the mandibular length and from about its midlength tapers very gradually anteriorly to an extremely small symphysis that is confined almost entirely to an anterolat- erally beveled surface at the end of the alveolar shelf. Posteriorly the dentary tapers to a long, acuminate process that lies in a groove on the lateral surface of the coronoid eminence of the surangular. The lightly built alveolar shelf has a maximum of 32 tooth positions. All the teeth are strongly recurved, sharply point- ed, have slightly serrated anterior cutting edges, and exhibit only modest variation in size. They appear to increase in length slightly to about the fifth or sixth tooth, then gradually decrease to the posterior end of the series, with the last six teeth being considerably smaller than the anteriormost teeth. In lateral view of the jaw 2001 Reisz and Berman — Late Permian Varanopseid from Russia 127 the splenial has a very splintlike exposure along the central portion of the ventral margin, with its dorsal margin contacting nearly equal lengths of the dentary and angular. The splenial is the dominant element on the medial surface of the jaw. Its dorsal margin, sheathing the lateral surface of the central portion of the alveolar shelf, gradually tapers anteriorly to the ventral margin of the jaw, to end as a slender, splintlike process at the level of the third tooth position and well short of the symphysis. The angular is exposed as a large, elongate element, occupying the posterior, ventral half of the lateral surface of the jaw. Posteriorly it contacts the ventral margin of the surangular, whereas anteriorly it sharply wedges between the dorsal dentary and the ventral splenial. From its lateral exposure the angular wraps around the ventral margin of the jaw to a narrow, medial exposure below the posterior portion of the prearticular. As in other varanopseids, there is no development of a ventral keel of the angular; but rather it has a rounded ventral margin that is smoothly continuous with the curvature of the rest of the jaw (Fig. 3). All but a small anterior portion of the lateral wall of the adductor fossa is formed by the surangular, the thickened, dorsal margin of which rises gradually anteriorly into a low, angular, coronoid eminence. The posterior coronoid is visible only in medial view of the jaw. Posteriorly it forms a short, narrow strip bordering the anterodorsal margin of the adductor fossa as it overlaps the medial surface of the surangular. Anteriorly the posterior coronoid narrows to a thin splint of bone which sheaths the posterior end of the medial surface of the alveolar shelf. There is no evidence of an anterior coronoid, and, although this is unusual among Permo-Carboniferous synapsids, the absence may be related to the unusually slender proportions of the mandible. Unfortu- nately, the mandible of Mycterosaurus is too poorly preserved in this area to allow comparison. Only in the holotype PIN 158/1 is the prearticular visible (Fig. 1), but most of its posterior extent is either poorly preserved or lost. What remains indicates a narrow strip for most of its extent along the medial or ventral border of the adductor fossa, widening considerably anteriorly before wedging sharply between the posterior coronoid and the splenial. The articular is exposed only partially in PIN 3706/11 and indicates the presence of a modest retroarticular process. Discussion The Varanopseidae is a clade of small to medium-sized synapsids that can now be characterized by a relatively large number of autapomorphic skeletal features which are associated usually with highly predaceous, lightly built faunivores (Dilkes and Reisz, 1996; Reisz et ah, 1998; and below). The family was erected by Romer and Price (1940) and originally contained two taxa, Varanops brevirostris and Aerosaurus greenleorum, which were collected from single, geographically and stratigraphically widely separated sites: the former from the well-known Early Permian (middle Leonardian) Cacops bonebed in the lowermost level of the un- divided Clear Fork Group of Hentz (1988; revised from Arroyo Formation, Clear Fork Group, of Romer, 1974), Baylor County, north-central Texas, and the latter from the Late Pennsylvanian (Late Gzelian) Cutler/ Abo Formation in El Cobre Canyon, Rio Arriba County, north-central New Mexico. Subsequent work has increased dramatically the list of known varanopseids to include Aerosaurus wellesi (Langston and Reisz, 1981) from the Early Permian (Wolfcampian) Cutler/ Abo Formation, Arroyo de Agua, Rio Arriba County, New Mexico, Mycterosaurus 128 Annals of Carnegie Museum VOL. 70 longiceps (Williston, 1915; Berman and Reisz, 1982) from the Early Permian (early Leonardian) Waggoner Ranch of the Wichita Group of Hentz (1988; revised from Clyde Formation, Wichita Group, of Romer, 1974), Mitchell Creek, Baylor County, Texas, Varanodon agilis (Olson, 1965) from the Late Permian (Guadal- upian) Chickasha Formation, Blaine County, Oklahoma, and Elliotsmithia longi- ceps (Broom, 1937; Dilkes and Reisz, 1996; Reisz et ak, 1998) from the Late Permian Tapinocephalus Assemblage Zone, Abrahamskraal, Western Cape Prov- ince, South Africa. Several taxa have been placed tentatively in Varanopseidae (Reisz, 1986), including Nitosaurus jacksonorum Romer and Price (1940), Myc- terosaurus smithae Lewis and Vaughn (1965), Milosaurus mccordi (DeMar, 1970), and Ruthiromia elcobriensis Eberth and Brinkman (1983). The cranial anatomy of Mesenosaurus provides overwhelming evidence of its varanopseid affinities in possessing the following synapomorphies: 1) dorsoventral expanded temporal fenestra occupies most of the height of temporal region, re- sulting in narrow subtemporal bar (some caseids show a similar condition); 2) marginal dentition is composed of strongly recurved, laterally compressed, sharply pointed teeth with fore and aft cutting edges restricted to the distal half of the tooth; 3) well-developed premaxillary subnarial shelf whose external surface is broadly rounded in transverse section; 4) anterior median process of the parietal extends into the supraorbital region of the skull table; 5) small postfrontal is bordered medially by a narrow, posterior process of the frontal; 6) tabular is reduced to a small, narrow element that contacts the medial margin of the pos- terolateral wing of the parietal; 7) absence of a medial, occipital flange of the posterior margin of the squamosal that covers the posterior margin of quadrate; 8) parasphenoid plate is broad throughout its length and the basisphenoidal tubera are winglike and extend far laterally and posteriorly from the base of the cultri- form process; and 9) a very prominent, nodular or tubercularlike ornamentation is present on the orbital margins of the prefrontal, postorbital, and jugal. The ninth varanopseid character listed above deserves some explanation, as it is so strikingly pronounced in the well-preserved skulls of Mesenosaurus that Efremov (1938) originally considered it to be an autapomorphy of the genus. Careful examination, however, reveals the presence of this feature in other var- anopseids in which the orbital bones are sufficiently preserved, despite in most instances severe damage due to over preparation. As examples, the holotypes of Mycterosaurus longiceps and Varanops brevirostris exhibit evidence of the same orbital-rim ornamentation found in Mesenosaurus. Elliotsmithia also exhibits the presence of this type of ornamentation on the same three orbital bones, although not commented on in the latest descriptions (Dilkes and Reisz, 1996; Reisz et ak, 1998). Although the relevant regions of the prefrontal and postorbital in the ho- lotype of Aerosaurus wellesi are too damaged to determine the presence of this feature, the well-preserved jugal exhibits the MesenosaurusAi]LQ ornamentation. Therefore, it is likely that the unusual pattern of ornamentation so well exempli- fied in Mesenosaurus actually diagnoses all varanopseids recognized to date. Mesenosaurus exhibits a number of cranial features recorded previously only in Mycterosaurus. Two of the most unusual of these are in the distinctive mor- phology of the maxilla, as they mimic the therapsid pattern in this structure: 1) the massive, anteroposteriorly broad dorsal lamina contacts the prefrontal, pre- venting contact between the nasal and lacrimal and reducing the lateral exposure of the lacrimal to a small, subrectangular area; and 2) the dorsal lamina forms a thick, deeply striated, abutment contact with the nasal. In both genera the pre- 2001 Reisz and Berman — Late Permian Varanopseid from Russia 129 frontal appears to be larger than in other varanopseids, with a well-developed ventral orbital process that nearly excludes the lacrimal from the orbital rim before continuing medially to it to contact the dorsal surface of the palatine. In both Mesenosaurus and Mycterosaurus the paroccipital process of the opisthotic is short and rodlike and, therefore, in sharp contrast to that in other varanopseids, where it is moderately to greatly expanded dorsoventrally into a bladelike struc- ture. Outgroup comparison with caseasaurs and ophiacodonts indicates that the latter state may be primitive for the family, but this is not certain. Although there are no well-preserved postcranial skeletons of Mesenosaurus in the collections of the Paleontological Institute of the Russian Academy of Sciences, a privately owned specimen reveals a number of features which, as far as known, occur also only in Mycterosaurus among the varanopseids. These include the presence of a modest lateral excavation at the base of the neural arches, two subequally devel- oped sacral vertebrae, and the absence of a supraglenoid foramen. Ongoing col- lecting from Mezen has already yielded new postcranial materials, preparation of which has just been initiated. Mesenosaurus and Mycterosaurus possess a number of features which appear in the derived state in other varanopseids, the most striking of which occur in the temporal region. In ElUotsmithia, Varanops, Aerosaurus, and Varanodon the tem- poral fenestra is expanded posteroventrally, resulting in an unusually large, rough- ly triangular- shaped opening. Possibly associated with this feature is a strong anterodorsal inclination of the occipital surface of the skull so that the braincase and the closely associated quadrate and quadrate process of the pterygoid are uniquely positioned relatively far anteriorly to occupy a position well into the region medial to the temporal fenestra. A consequence of this change in skull proportions is the reduction of the adductor chamber, the space normally formed between the parietal and postorbital bones dorsally, the quadrate ramus of the pterygoid medially, the quadrate and squamosal posteriorly, and the zygomatic arch laterally. Two derived features of Aerosaurus, Varanops, and Varanodon which appear in the primitive state in Mesenosaurus and Mycterosaurus include the presence of a well-developed retroarticular process of the articular and the inclusion of the quadratojugal into the temporal fenestra. Unfortunately, in El- liotsmithia the posterior region of the mandible is not sufficiently preserved to indicate the presence or absence of a retroarticular process. The presence of a long posterior process of the jugal which contacts the squamosal to exclude the quadratojugal from the temporal fenestra in Elliotsmithia, as it does in Myctero- saurus and Mesenosaurus, is the one notable contradiction to the dichotomy of the varanopseids proposed here. Conclusions In the recent study of the varanopseid Elliotsmithia longiceps by Reisz et al. (1998) a cladistic analysis of the phylogenetic intrarelationships of the family was presented that demonstrated a basal dichotomy, and, although two distinct clades were recognized, subfamilial units were not formally proposed for either. The cladogram of Varanopseidae phytogeny presented by Reisz et al. (1998) is repro- duced here (Fig. 7), as it agrees with the comparative, anatomical data presented above and therefore supports the following generic relationships and subfamial assignments: 1) Mesenosaurus is a member of the family Varanopseidae as a sister taxon to Mycterosaurus and together they form a clade which is assigned here 130 Annals of Carnegie Museum VOL. 70 PALEOZOIC carboniferous] PERMIAN IN) Cn '< 3 S> ■0"'i w a S) ‘ProtocI epsydrops m c ■O '< o o <1 A V 0 1.5 ■ • 1.4 • ■ LAVraUo M2/ 1.3 ■ 1.2 J LM’ ratio M/2 !.3 1.. Fig. 4. — Scatter diagrams of LAV ratios of <\Eumyarion medium^ Sansan; V Eumyarion latior, Aliveri; ElEumyarion bifidus, Sandelzhausen; oEumyarion bifidus, Puttenhausen; \>Eumyarion montanus, Ke- sekoy; AEumyarion carbonicus, Harami 1; oEumyarion aff. carbonicus, Harami 3; * Prokanisamys arifi, H~GSP 116; ^Prokanisamys arifi, H-GSP 8311; W Prokanisamys major n. sp., H-GSP 8224; •^Prokanisamys major, H-GSP 8114; ^Prokanisamys benjavuni, Li; ^Prokanisamys benjavuni, H- GSP 8114; uProkanisamys benjavuni, H-GSP 8424; AProkanisamys kowalskii, Z 113. Prokanisamys major, new species (Fig. 5A-L) Derivatio nominis. — “major” because the cheek teeth are large. Holotype.—yP sin no. 4522, 21.0M6.2 (Fig. 5C~D). Type Locality. — H-GSP 8114. Type Level. — Lower Manchar Formation (Middle Miocene). Occurrences. — Sind, Lower Manchar Formation, Sehwan section, localities H- GSP 8106, 8114, 8227, 8224, 8424, 8214.8425,8426, 8427; Potwar Plateau: Y- GSP 591, 592, 642, 501, 589, 491, 640, 641, 496, 634 (Flynn, 1986). Diagnosis. — Prokanisamys major represents a large Prokanisamys with the main cusps clearly distinguishable; the lophs are (in fresh specimens) lower than Table 1 . — Summary of the major morphological differences between Eumyarion and Prokanisamys species. Eumyarion Prokanisamys Ml and M^ 1) Metalophule transverse connecting to hypo- cone 2) Anterior arm of protocone often present in M® 3) Metacone inflated Ml and M2 1) Posterior arm of protoconid and mesolophid usually present 2) Free ending posterior arm of the hypoconid usually present 3) Entoconid inflated Ml and M^ 1) Metalophule posteriorly connected to pos- teroloph 2) Anterior arm of protocone never present 3) Metacone not inflated Ml and M2 1) Never two ridges in the mesosinusid present (never a posterior arm of protoconid and a mesolophid) 2) Free ending posterior arm of hypoconid never present 3) Entoconid not inflated 150 Annals of Carnegie Museum VOL. 70 Fig, 5. — Prokanisamys major from H-GSP 8114. A,B- Occlusal, labial views M', H-GSP 8114/4521, C,D. Occlusal, labial views M’, H-GSP 8114/45223. E,F. Occlusal, labial views M^, H-GSP 8114/ 4557. G,H. Occlusal, labial views M„ H-GSP 8114/4582. 1,J. Occlusal, labial views M3, H-GSP 8114/ 4612. K,L. Occlusal, labial views M3, H-GSP 8114/4614. IProkanisamys sp. A from H-GSP 8107(a). M,N. Occlusal, labial views M„ H-GSP 8107a/4015. 0,P. Occlusal, labial views M3, H-GSP 8107/ 4068, Letters underlined: original is from the right side. 2001 Wessels and de Bruijn— "Miocene Rhizomyidae from Pakistan 151 the cusps and the M3 is shorter than the Mj. The anterolophid is not or poorly connected to the anteroconid. The metalophulid, hypolophulid, protolophule and metalophule are short. Differential Diagnosis,- — Prokanisamys major specimens are morphologically similar to P. arifi, larger (although a slight overlap may occur) and relatively lower crowned. Prokanisamys major differs from P. benjavuni in having larger first molars, smaller third molars, the and are relatively narrower, the molars are more cuspate (especially the lower), the has a protosinus and the labial cusps in the lower molars lack an antero-labial ridge. Prokanisamys major specimens are morphologically close to P. kowalskii. Al- though the size-range of P. kowalskii and P. major overlap, except for the and the third molars and the largest P. major specimens, P. major is considered to represent a different species because the mean size of P. kowalskii specimens is larger than the mean sizes of P. major from the older localities. Furthermore the of P. kowalskii is much smaller than in P. major whereas the is larger and the M3 much larger. Material and Measurements 3 and Appendix 3, Appendices 7--11. Description of the Type Material.— M\ The cusps are bulbous; height of molars moderate; anterocone short and crescent shaped; the two anterocone cusps are slightly separated; the lingual cusps are slightly compressed transversally and a protosinus is present; the protocone is weak or poorly connected to protolophule or longitudinal crest; the mesoloph is short and transverse or directed slightly posteriorly; the postero-sinus is present and persists in worn specimens. In the holotype, a small cusp is present on the antero-lingual base of the hypocone. The slightly curved sinus is directed forwards. M2. The anterior part of this molar is missing. The cusps are bulbous; the height is moderate; lingual cusps are slightly compressed transversally; the longitudinal crest is oblique; the mesoloph short; the postero-sinus is less developed than in the M* and disappears through wear; the sinus is slightly curved forwards and does not extend anteriorly of the protolophule, M^. The cusps are clearly defined in unworn specimens; the labial branch of the anteroloph is preserved in fresh specimens only; the mesoloph is short; the posteroloph is well developed; the metacone is clearly visible; the paracone, the longitudinal crest and the hypocone are connected in advanced wear stages. Mj. The cusps are bulbous; the lophs are lower than the cusps; the anteroconid is simple and connected to the metaconid by a short but high lingual branch of the anterolophid; the anterolophulid is poorly or weakly developed in all specimens and not connected to the anteroconid. The metalophulid is always short, transverse or slightly backwards directed and poorly connected to protoconid. The longitudinal crest is directed obliquely towards the entoconid; the hypolophulid is short. The hypoconid is in three specimens weakly connected to the longitudinal crest. The mesolophid is short; a short ridge on the postero-lingual edge of the metaconid descends to the base of the mesosinusid; the posterolophid is strong; the sinusid directed backwards. Mj. The cusps are bulbous; the lophs are lower than the cusps; the lingual branch of the anterolophid is absent and Ae labial branch is short; the longitudinal crest is obliquely directed; the metalophulid, hypolophulid and mesolophid are short and transverse. A short ridge on the postero-lingual edge of the metaconid descends to the base of the mesosinusid; the posterolophid is strong; the sinusid is transverse. M3, The cusps are bulbous; the lophs are lower than cusps; the very short branches of the antero- lophid disappear with wear; the metalophulid and hypolophulid are short; the longitudinal crest is strong and obliquely directed; the mesolophid is either of medium length or reaches the lingual edge; a clear postero-sinusid is present; the sinusid is transverse or directed posteriorly. Comparisons and Discussion.— Prokanisamys major and P. arifi differ in size, but an overlap is present (Fig. 3). P. major is relatively lower crowned and slightly more robust than P. arifi. The cuspate nature of especially the lower molars of Prokanisamys major distinguishes these from specimens of P. benjavuni. The labial cusps in the lower molars of Prokanisamys major do not have an antero- 152 Annals of Carnegie Museum VOL. 70 labial ridge and the upper molars do not have a ridge on the antero-lingual side of the lingual cusps. The of P. major has a protosinus, which is lacking in P. benjavuni. The and of P. major and P. benjavuni are more difficult to distinguish than the lower molars because they have a similar degree of hypso- donty. Moreover, the variation in surface area is large due to differences between wear stages (Appendix 9). P. major is lower crowned than P. kowalskii (Lindsay, 1996). P. major differs from P. kowalskii in size, and in having a more obliquely directed longitudinal crest, a short and transverse protolophule in the M‘; the protocone is often poorly connected to the longitudinal crest, and anterolophulid and metalophulid (Mj) are weakly developed. Prokanisamys major differs from Kanisamys indicus in being smaller, less hyp- sodont and in having the cricetid bunodont pattern, Prokanisamys major is represented by 71 specimens from seven localities. In the younger assemblages the molars are slightly larger and the upper molars are relatively higher, the labial branch of the anteroloph (M^) is stronger and may end in a small cuspule, and the posterolophid is stronger. Prokanisamys major and P. arifi share a similar morphology, but P. major is larger and lower crowned and is therefore not considered to be a direct descendant from P. arifi. Also P. kowalskii and P. major have a similar morphology, but P. major is not considered to be a descendant of P. kowalskii because of differences in relative sizes of the cheek teeth. Most cheek teeth of P. kowalskii are larger than those of the geologically oldest specimens of P. major (Appendices 3, 9 and 10). Prokanisamys benjavuni (Mein and Ginsburg, 1985) (Fig. 6A-P) Type Locality. — Li basin, Thailand. Type Level. — Miocene. Diagnosis. — “Espece du genre Prokanisamys differant de Prokanisamys arifi par des M^ plus petites et des M^ plus grandes, une hypsodontie plus forte tout en restant moderee, des surfaces occlusales peu concaves par suite du develop- pement de la lophodontie. Grande reduction de L anterolophide qui disparait meme complete ment sur M^. Hypolophulides iamais oblique vers V arriere.” (Mein and Ginsburg, 1997). Addition to Diagnosis. — The anterocone and protocone are strongly connected by the lingual branch of the anteroloph; a protosinus is absent. In Prokanisamys benjavuni the M^ is relatively broader, the M^ larger and the M^ and M3 are much larger than in P. arifi. The Mj has a narrow anteroconid-complex. The cheek teeth are more lophate than P. arifi. Occurrences. — Li Basin, Thailand (Mein and Ginsburg, 1985, 1997); Bugti area, in levels 5 and 6 (Welcomme and Ginsburg, 1997); Sind, Lower Manchar Formation, Sehwan section, localities H-GSP 8114^2, 8114, 8424, 8426. Material and Measurements. — Figure 3 and Appendix 4, Appendices 7-11. Comparisons and Discussion. — Although the variation in size (especially M3) is large in the Li material. Mein and Ginsburg concluded that only one species is present. They considered the size variation to be the result of wear: hypsodont molars become shorter and broader when worn. The specimens from Pakistan are on average slightly larger than those from 2001 Wessels and de Bruijn — Miocene Rhizomyidae from Pakistan 153 Fig. 6. — Prokanisamys benjavuni from H-GSP 8114. A,B- Occlusal, labial views M’, H-GSP 8114/ 4523. C,D. Occlusal, labial views M^, H-GSP 8114/4543. E,F. Occlusal, labial views M^, H-GSP 81 14/ 4554. G,H. Occlusal, labial views M', H-GSP 8114/4524. I,J. Occlusal, labial views M^, H-GSP 8114/ 4542. K,L. Occlusal, labial views M,, H-GSP 8114/4577. M,N. Occlusal, labial views M2, H-GSP 81 14/4593. 0,R Occlusal, labial views M3, H-GSP 8114/4618. Letters underlined: original is from the right side. 154 Annals of Carnegie Museum VOL. 70 Fig, 7. — Kanisamys indicus from H-GSP 8224. A,B- Occlusal, labial views M\ H-GSP 8224/4502. C,D. Occlusal, labial views H-GSP 8224/4517. E,F. Occlusal, labial views M^, H-GSP 8224/4524. G,H. Occlusal, labial views M„ H-GSP 8224/4542. I,J. Occlusal, labial views M2, H-GSP 8224/4556. K,L. Occlusal, labial views M3, H-GSP 8224/4575, Letters underlined: original is from the right side. 2001 Wessels and de Bruun — Miocene Rhizomyidae from Pakistan 155 Thailand. and are broader and the is relatively larger. The lower molars are slightly larger (M2 and M3) or within the upper range of the specimens from Li (Ml). The Pakistani specimens are morphologically similar to those of Thai- land, but the M3 specimens show a wider morphological variation. The roots of the specimens from both areas are the same. M^ has a broad lingual root with a deep groove in its center and the M^ has four roots. Flynn (1986) and Jacobs et al. (1989) have reported Prokanisamys benjavuni from some localities from the Potwar Plateau. However, the specimens figured in Jacobs et al. (1989) and casts thereof clearly represent Prokanisamys major. Mein and Ginsburg (1997) already observed important differences between the Thai and the Potwar Plateau material assigned to P. benjavuni, and correctly concluded that these do not belong to the same lineage. Prokanisamys benjavuni and Brachyrhizomys species (Potwar Plateau, nine Ma, Flynn, 1982a) have some characteristics in common: the shape of the M^ which lacks a protosinus (flexus or anterior sulcus in Flynn, 1982a), and the deep sinusid in the lower molars. Prokanisamys benjavuni is much smaller and only slightly hypsodont, but its cheek-teeth morphology suggests it could belong to the same lineage as Brachyrhizomys. If that is so, the separation between the Rhizomyinae and the Tachyoryctinae is older than previously thought. The ’sudden’ occurrence of several Brachyrhizomys species in the Nagri Formation and Dhok Pathan For- mation from the Potwar Plateau (Flynn, 1982a) could be interpreted as the result of a rapid diversification of this genus after its immigration from a more eastern region. ^Prokanisamys sp.A (Fig. 5M-P) Occurrences.— Sind, Lower Manchar Formation, Sehwan section; locality H- GSP 8107(a). Material and measurements .—Pigure 3 and Appendix 5, Appendices 7~11. Description of the Material. — The single M’ is severely damaged and shows a clear protosinus, but no further detail. The lower molars have low bulbous cusps. The lophs are lower than the cusps. The anteroconid of the Mj is connected to the metaconid by the low lingual branch of the anterolophid. On the M2 the anterolophid has two well-developed branches. The mesolophid is of medium length. Comparisons and Discussion.—This very large Prokanisamys is larger than most Prokanisamys major specimens from younger localities. The teeth are cus- pate and have low crowns, typical of Prokanisamys. IProkanisamys sp.A differs from P. kowalskii in having a more obliquely directed longitudinal crest (MO and in the interrupted protoconid-anteroconid connection (Mj). The transverse metal- ophulid is directed towards the anterior part of the protoconid, but not connected to it, 1 Prokanisamys sp.A differs from P. major in having a larger and higher M^ Ml. This species may be seen as a descendant of P. kowalskii (Lindsay, 1996). The teeth are within the size-range of that species but they seem to be more hypsodont and the Mi has a more anteriorly directed metalophulid (derived char- acters). The M1/M3 length-ratio is larger than 1.0, which is a Kanisamys charac- teristic. These few teeth with a cuspate cricetid dental pattern and a M1/M3 length- ratio as in Kanisamys seem to represent a new species. We refer this species as IProkanisamys sp.A because the material is considered insufficient as the basis for a formal name. 156 Annals of Carnegie Museum VOL. 70 Kanisamys Wood, 1937 Type Species. — Kanisamys indicus Wood, 1937. Other Species Included. — Kanisamys sivalensis Wood, 1937; Kanisamys nagrii Prasad, 1968; Kanisamys potwarensis Flynn, 1982«. Diagnosis. — “Small tachyoryctines with moderate to advanced hypsodonty and lophodonty; lower massetric crest strongly inclined; mesolophids become reduced through time but are distinct; incisors small with strongly rounded enamel; dentary depth shallow” (Flynn, 1982^). Kanisamys indicus Wood, 1937 (Fig. 7A-L) Type Locality. — South of Chinji (see Wood, 1937 and Black, 1972). Type Level. — Chinji Formation. Original Diagnosis. — “Smallest known Kanisamys, lophodont but with rec- ognizable cusps; crown height is moderate; mesolophid and hypolophid usually strong and separate in all lower molars; anterolophid of unworn M2_3 strong and with distinct labial arm; Incisor with two ridges” (Flynn, 1982a). Occurrences. — In many localities in the Chinji Formation: Potwar Plateau (Wood, 1937; Black, 1972; Dehm et ah, 1982; Jacobs et ah, 1989); Trans Indus, Baud Khel (Munthe, 1980); N.W.F.P, Banda Baud Shah (Wessels et ah, 1982). Sind H-GSP 8114, 8227, 8224, 8425, 8427, 8214 from the Lower Manchar For- mation. Material and Measurements. — Figure 3 and Appendix 6, Appendices 6-10. Comparisons and Discussion. — Kanisamys indicus cheek teeth are larger and more hypsodont than those of Prokanisamys major and P. benjavuni although some elements show a slight overlap (Fig. 3). The lophs are more prominent than in Prokanisamys major. The of K. indicus has, in contrast to P. benjavuni, a protosinus and a transverse or slightly forward-curved sinus, which never reaches the protolophule. Kanisamys indicus specimens from H-GSP 82.24 have higher lophs and are relatively higher-crowned than specimens from older localities (H- GSP 8228, 8114). The Miocene Record oe the Rhizomyidae in Pakistan The oldest occurrence of the Rhizomyidae is Prokanisamys kowalskii, from locality Z 113 (Chitarwata Formation, Zinda Pir Borne; Bowning et ah, 1993; Lindsay, 1996), a locality which has been assigned a tentative age of 20 Ma (Lindsay, 1996). P. arifi occurs in slightly younger assemblages in west-central Pakistan (Chitarwata Formation and Vihowa Formation, Zinda Pir Borne; Bown- ing et al., 1993; Lindsay, 1996) northern Pakistan (Murree Formation; de Bruijn et al., 1981) and southern Pakistan (Lower Manchar Formation; de Bruijn and Hussain, 1984; Fig. 8). The replacement of P. kowalskii by P. arifi in assemblages that are only some- what younger, and the occurrence of several other rhizomyid species in association with P. arifi in immediately successive sites suggests a rapid diversification of the Rhizomyidae during their early history (Fig. 8). IProkanisamys sp. A, which is known from locality H-GSP 8107(a) only, seems to be the descendant of P. kowalskii. The phylogenetic relationship between P. arifi, P. kowalskii and P. major is not clear. Prokanisamys benjavuni, which was originally described from the locality Li 2001 Wessels and de Bruijn — Miocene Rhizomyidae from Pakistan 157 Reconstructed sequence of localities Prokanisamys arifi Prokanisamys major Prokanisamys benjavuni IProkanisamys sp. A Kanisamys indicus H-GSP 82.14 H-GSP 82.24 H-GSP 84.27 H-GSP 82.27 H-GSP 8426 H-GSP 8425 H-GSP 8114 H-GSP 8424 H-GSP 8114a H-GSP 8106 H-GSP 8311 H-GSP 8209 H-GSP 8107(a) ' ■ Fig. 8. — Distribution chart of the Rhizomyidae from the Lower Manchar Formation in Sind. The localities are listed in stratigraphical order. in Thailand (Mein and Ginsburg, 1985; 1997), seems to be an immigrant into Pakistan from the east because all Pakistani records, including the oldest teeth known from level 6 in the Dera Bugti area (Welcomme and Ginsburg, 1997, and Welcomme et ah, 1997), show more derived characters than the ones from Li. The first record of P. benjavuni in Pakistan is therefore considered to be somewhat younger than the Li locality. Comparison of the rodent associations from level 6 with those from the base of the Manchar Formation in the Gaj and Sehwan areas (Wessels, 1996) suggests that an age of 16 to 17 million years is more probable for level 6 than the 18 million years suggested by Welcomme et al. (1997). The cheek teeth of Kanisamys indicus are larger and higher-crowned than those of P. kowalskii and P. arifi, so either could be ancestral to K. indicus. The Miocene record of the Rhizomyidae of Pakistan shows a mosaic pattern of appearances and extinctions with the exception of Kanisamys indicus and Kan- isamys nagrii, which probably are closely related and may belong to one lineage (Flynn, 1990:597). Ancestor-descendant relationships are not clear because grad- ual transitions between species are not documented and the ranges of species that are potentially members of one lineage (partly) overlap (Fig. 2; Flynn, 1982^, 1986). This peculiar pattern can only be explained by postulating migrations of allopatric species. The largest diversity among the Rhizomyidae in the Manchar Formation occurs in locality H-GSP 81.14. Here Prokanisamys arifi, P. major, and P. benjavuni, known from older levels, are associated with the newcomer Kanisamys indicus. Shortly after this rhizomyid optimum the number of species drops to two: Pro- kanisamys major and Kanisamys indicus. These two species share a range of about five million years (see also Jacobs et al., 1989; Flynn, 1990). 158 Annals of Carnegie Museum VOL. 70 The earliest adaptations to a fossorial lifestyle have been shown to occur in Brachyrhizomys (Flynn, 1982^:; 1990) from a level of the Potwar Plateau in Pak= istan that has been assigned an age of nine Ma. Prokanisamys benjavuni, with its incipiently hypsodont cheek teeth, without a protosinus, and lower molars with deep sinusids is known from its dentition only, but this dentition shows adaptations characterizing other fossorial rodent subfamilies (Spalacinae, Anom- alomyinae). It is therefore suggested that the separation between the Rhizomyinae and the Tachyoryctinae occurred much earlier than hitherto recognized. The Potential of the Rhizomyidae for Biostratigraphy Figure 8 shows the Rhizomyidae from the Lower Manchar Formation and aL lows the recognition of two assemblage zones. The lower zone is characterized by the presence of Prokanisamys arifi. In the upper part of this zone P. arifi occurs associated with P. benjavuni, P. major and/or K. indicus. The co-occur- rence of P. major and K. indicus and the absence of P. arifi and P. benjavuni characterize the upper zone. The occurrences of several Rhizomyidae species in the succession of the Potwar Plateau (Jacobs et aL, 1989) allows the recognition of three zones. In the lower zone Prokanisamys major and Kanisamys indicus co-occur, P. arifi and P. ben- javuni are absent (same as the upper zone from the Lower Manchar Formation). The second zone is characterized by the replacement of Kanisamys indicus by its successor K. nagrii and the absence of P. major. The third zone in the Potwar Plateau is characterized by the appearance of several species of Brachyrhizomys. It seems that a foursome division of the Pakistani Miocene is possible on the basis of the Rhizomyidae, but further information is needed to define formal biozones. The Age of the Gaj and Sehwan Assemblages Since dependable palaeomagnetic data on the Gaj and the very condensed Seh- wan sections are not available, age estimates of our assemblages are based on biostratigraphical correlation with assemblages from the Potwar region and the Sulaiman range which are tied to palaeomagnetic data. Difficulties in recognizing similarities in the composition of assemblages are as follows: differences in taxonomic identification and interpretation by different authors; absence of information on the intraspecific variation (either due to stasis or to lack of description of this variation); variation differences due to geograph- ical distribution, preservation, sedimentation (sorting) and/or sampling method. Correlations of the Sehwan and Gaj faunas based on the preliminary faunal lists by de Bruijn et al. (1984) have been suggested by several authors (i.e., Bemor et al., 1988; Lindsay and Downs, 1998; Jacobs et al., 1989). More precise cor- relations will not be possible before all the material will have been correctly identified and the intraspecific variation known. Wessels (1996) gave conservative age estimates, based on the detailed comparison of the Myocricetodontinae from the Potwar region with those from Sind. The Sehwan localities were then consid- ered to have ages between 13.7 and 16.2 Ma, the upper Gaj locality younger than 13.7, and the lower Gaj localities older than 16.2 Ma but younger than the Murree assemblage. Lindsay (1996) concluded, on basis of the Rhizomyidae, that the locality from the base of the Murree Formation near Banda Daud Shah is slightly younger than 2001 Wessels and de Bruijn— Miocene Rhizomyidae from Pakistan 159 locality Z 113 from the Pinda Zir Dome (Sulaiman Range), which is dated at approximately 20 Ma. We follow Lindsay and consider the lower Gaj localities to have ages between 16.2 and 19.5 Ma. Acknowledgments We thank the Director General of the Geological Survey of Pakistan (GSP) for giving permission and providing facilities for fieldwork. Dr. S. Mahmood Raza and Mr. M, Arif from the Geological Survey of Pakistan have organized the fieldwork. Drs. Sjef Leinders, Hans Thewissen, Ed Cousin, M. Arif, Hassan Shaheed and Farid Ahmed Uddin have helped to collect the material. Prof. P. de Boer provided the information on the Sehwan Sharif sections. Dr. Flynn and Dr. Lindsay are acknowledged for sending casts. Prof. Dr. Fahlbusch kindly provided data on Eumyarion. Dr. Lindsay and Dr. Flynn provided constructive reviews. This paper is part of Howard University- Geological Survey of Pakistan project “Cenozoic Mammals of Pakistan” directed by Dr. S. Taseer Hussain of the Howard University, Washington, D.C., U.S.A. The project was supported by several grants from the Smithsonian Institution and the U.S. National Science Foundation (grant numbers 20868200 and DEB-8003601, respectively) awarded to Dr. S. Taseer Hussain. W. den Hartog made the S.E.M. images and J. Luteyn made the plates. Literature Cited Bernor, R. L., L. J. Flynn, T Harrison, S. T Hussain, and J. Kelley. 1988. Dionysopithecus from southern Pakistan and the biochronology and biogeography of early Eurasian catarrhines. Journal of Human Evolution, 17:339--358, Black, C. C. 1972. Review of Fossil rodents from the Neogene Siwalik beds of India and Pakistan. Paleontology, 15 (2): 23 8-266. Bruijn, H. de, and S. T. Hussain. 1984. The succession of rodent faunas from the Lower Manchar Formation, Southern Pakistan and its relevance for the biostratigraphy of mediterranean Miocene. Paleobiologie continentale, XIV(2): 19 1-202. — . 1985. Thryonomyidae from the Lower Manchar Formation of Sind, Pakistan. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, B, 88(2): 155-166. Bruijn, H. de, and G. Sara^. 1991. Early Miocene rodent faunas from the western Mediterranean area. Part I. The genus Eumyarion. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, B, 94(1): 1-36. Bruijn, H. de, E. Boon, and S. T. Hussain. 1989. Evolutionary trends in Sayimys (Ctenodactylidae, Rodentia) from the Lower Manchar Formation (Sind, Pakistan). Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, B, 92(3): 191-214. Bruijn, H. de, S. T. Hussain, and J. J. M. Leinders. 1981. Fossil rodents from the Murree Formation near Banda Daud Shah, Kohat, Pakistan. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, B, 84(l):71-99. Carleton, M. D,, and G. G. Musser. 1984. Muroid rodents. Pp. 289-379, in Orders and families of recent mammals of the World (S. Anderson and J. K. Jones, eds.). John Wiley, New York. Chaline, j. P. Mein, and F. Petter. 1977. Les grandes lignes d’une classification evolutive des Muroidea. Mammalia, 41:245-252. Dehm, R., L. L. Jacobs, W. Wessels, H. de Bruijn, and S. T. Hussain. 1982. Fossil rodents from the type area of the Chinji Formation, Siwalik group, Pakistan. Proceedings of the Koninklijke Ned- erlandse Akademie van Wetenschappen, B, 85(3):259-263. Downing, K. E, E. H. Lindsay, W R. Downs, and S. E. Speuer. 1993. Lithostratigraphy and verte- brate biostratigraphy of the early Miocene Himalayan Foreland, Zinda Pir Dome, Pakistan. Sed- imentary Geology, 87:25-37. Fahlbusch, V. 1964. Die cricetiden (Mammalia) der Oberen Siisswasser-Molasse bayerns. Abhan- dlungen der Bayerische Akademie der Wissenschaften, Mathematisch-Naturwissenschaftliche Klasse, Neue Folge, 118:1-136. Flynn, L. J. 1982^1. Systematic revision of Siwalik Rhizomyidae (Rodentia). Geobios, 15(3):327- 389. — — . 1982Z?. A revision of fossil rhizomyid rodents from northern India and their relation to a rhizomyid biochronology of Pakistan. Geobios, 15:583-588. . 1986. Species longevity, stasis and stairsteps in rhizomyid rodents. Pp. 273-285, in Verte- brates, phylogeny and philosophy (M. Flanagan and J. A. Lillegraven, eds.). Contributions in Ge- ology, University of Wyoming, Special Paper, 3. 160 Annals of Carnegie Museum VOL. 70 . 1990. The natural history of Rhizomyid rodents. Pp. 155-183, in Evolution of subterranean mammals at the organismal and molecular levels (E. Nevo, and O. A. Reig, eds.). A. R. Liss, Inc., New York. Jacobs, L. L., L. J. Flynn, and W. R. Downs. 1989. Neogene rodents of Southern Asia. Pp. 157- 177, in Papers on fossil rodents in honor of Albert E. Wood (C. C. Black and M. R. Dawson, eds.). Los Angeles County Museum Sciences series, 33. Kalthoff, D. C. 1999. Die Schmeltzmikrostruktur in den Incisiven der hamsterartigen Nagetiere und anderer Myomorpha (Rodentia, Mammalia). Ph.D. Thesis, Bonn, Germany. Kleinhofmeijer, G., and H. de Bruijn. 1988. The mammals from Aliveri (island of Evia, Greece). Part 8. The Cricetidae. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschap- pen, B, 91(2): 185-204. Lartet, E. 1851. Notice sur la colline de Sansan. 1 volume. J. -A. Portes, Auch. Lindsay, E. H. 1996. A new eumyarionine cricetid from Pakistan. Acta zoologica Cracoviensis, 39(1): 279-288. Lindsay, E. H., and W. R. Downs. 1998. Cricetid rodents from Miocene Deposits of Pakistan. 1. Pp. 35-47, in 3e Geosas workshop, Pakistan (Ghaznov, S. M. Raza and M. T. Hasan, eds.). Mein, R, and M. Freudenthal. 1971. Les cricetidae de Vieux-CoUonges. Partie I. Scripta Geologica, 5. Mein, P, and L. Ginsburg. 1985. Les rongeurs miocenes de Li (Thailande). Comptes Rendus de 1’ Academie des Sciences, Paris, t. 301, II, 19:1369-1374. ■ — . 1997. Les mammiferes du gisement Miocene inferieur de Li Mae Long, Thailande: syste- matique, biostratigraphie et paleoenvironment. Geodiversitas, 19(4):783-844. Miller, G. S., and J. W. Gidley. 1918. Synopsis of supergeneric groups of rodents. Journal of the Washington Academy of Sciences, 8:431-448. Munthe, j. 1980. Rodents of the Miocene Daud Khel Local fauna, Mianwali District, Pakistan. Part II. Sciuridae, Gliridae, Ctenodactylidae and Rhizomyidae. Milwaukee Public Museum, Contri- butions in Biology and Geology, 34:1-36. Nevo, A. 1999. Mosaic evolution of subterranean mammals. Regression, progression and global convergence. Oxford University Press, Oxford, United Kingdom. Nowak, R. M. 1991. Walker’s mammals of the World. Fifth edition. J. Hopkins University Press, London, United Kingdom. Prasad, K. N. 1968. The vertebrate fauna from the Siwalik beds of Haritalyangar, Himachal Pradesh, India. Memoirs of the Geological Survey of India, Palaeontologia Indica, 39:1-56. ScHAUB, S. AND H. Zapfe. 1953. Die Fauna der miozanen Spaltenfullung von Neudorf a. d. March. Simplicidentata. Sitzungsbereich der Osterreichische Akademie der Wissenschaften, Matehema- tisch-Naturwissenschaftliche Klasse, 162(1): 18 1-2 15. Unay-Bayraktar, E. 1989. Rodents from the middle Oligocene of Turkish Thrace. Utrecht Micro- paleontological Bulletins, Special Publication, 5. Welcomme, J-P, P-O. Antoine, E Duranthon, P. Mein, and L. Ginsburg. 1997. Nouvelles decou- vertes de Vertebres miocenes dans le synclinal de Dera Bugti (Balouchistan, Pakistan). Comptes rendus de 1’ Academie des Sciences, Paris, 325:531-536. Welcomme, J-R, and L. Ginsburg. 1997. Mise en evidence de 1’ Oligocene sur le territoire des Bugti (Balouchistan, Pakistan). Comptes rendus de TAcademie des Sciences, Paris, 325:999-1004. Wu, W. 1982. Die Cricetiden (Mammalia, Rodentia) aus der Oberen Siisswasser-Molasse von Put- tenhausen (Niederbayern). Zitteliana, 9:37-80 Wessels, W. 1996. Myocricetodontinae from the Miocene of Pakistan. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, 99(3-4):253-312. Wessels, W, H. de Bruijn, S. T. Hussain, and J. J. M. Leinders. 1982. Fossil rodents from the Chinji Formation, Banda Daud Shah, Kohat, Pakistan. Proceedings of the Koninklijke Nederlandse Aka- demie van Wetenschappen, B, 85(3):337-364. Wessels, W, E. Unay, and H. Tobien. 1987. Correlation of some Miocene faunas from Northern Africa, Turkey and Pakistan by means of Myocricetodontinae. Proceedings of the Koninklijke Nederlandse Akademie van Wetenschappen, B, 90(l):65-82. Wood, A. E. 1937. Fossil rodents from the Siwalik beds of India. American Journal of Sciences, New Haven, 36:64-76. 2001 Wessels and de Bruijn — Miocene Rhizomyidae from Pakistan 161 Appendix 1 Material and measurements o/ Prokanisamys arifi. Prokanisamys arifi Locality Length Width N Mean Range N/N Mean Range H^GSP 8426 M* 1 20.4 1/1 15.6 H-GSP 8114 M' 0 0/0 H-GSP 8424 M' 0 0/0 H-GSP 8114a M' 1 0/0 H-GSP 8106 M> 4 19.200 1/2 14.500 14.3-14.7 H-GSP 8311 Ml 9 19.750 18.0-20.9 5/6 14.967 14.6-15.6 H-GSP 8107 (a) M’ 1 0/0 H-GSP 8426 M2 0 0/0 H-GSP 8114 M2 0 0/0 H-GSP 8424 M2 0 0/0 H-GSP 8114a M2 9 14.857 14.0-15.9 7/9 14.611 13.1-16.5 H-GSP 8106 M2 5 14.650 14.1-15.2 4/3 14.300 13.6-14.9 H-GSP 8311 M2 14 14.920 14.3-15.6 10/10 14.240 13.4-15.3 H-GSP 8107 (a) M2 5 15.125 14.4-15.8 3/3 14.500 13.1-15.3 H-GSP 8426 M2 2 12.350 11.2-13.5 2/2 13.050 12.4-13.7 H-GSP 8114 M2 4 12.600 12.0-13.2 2/2 12.950 12.9-13.0 H-GSP 8424 M2 6 11.933 11.5-12.5 3/3 12.200 11.8-12.6 H-GSP 8114a M2 3 12.700 11.9-13.7 3/3 13.600 12.9-14.7 H-GSP 8106 M3 5 12.450 11.8-13.4 4/4 12.675 11.9-13.7 H-GSP 8311 M2 16 11.486 10.3-12.4 14/14 12.200 11.5-13.3 H-GSP 8107 (a) M2 3 11.900 11.5-12.5 3/2 12.600 12.4-12.8 H-GSP 8426 M, 3 17.550 17.0-18.1 2/2 12.900 12.4-13.4 H-GSP 8114 M, 2 18.200 1/1 10.800 H-GSP 8424 M, 1 17.000 1/1 11.200 H-GSP 8114a M, 5 18.133 17.7-18.6 3/4 12.650 11.2-13.4 H-GSP 8106 M, 6 18.333 17.8-19.0 3/3 12.700 12.5-12.8 H-GSP 8311 M, 15 17.510 16.8-18.2 10/9 11.944 10.9-12.9 H-GSP 8107 (a) M, 7 18.450 18.1-18.7 4/5 12.660 12.0-13.2 H-GSP 8426 M, 1 15.000 1/0 H-GSP 8114 0 0/0 H-GSP 8424 M2 1 0/0 H-GSP 8114a M2 6 16.700 1/2 15.050 14.8-15.3 H-GSP 8106 M2 8 15.750 15.1-16.1 6/5 14.020 13.2-14.6 H-GSP 8311 M2 18 15.762 14.4-17.4 13/14 14.136 13.1-15.9 H-GSP 8107 (a) M2 11 16.4 15.6-17.5 6/7 14.471 13.5-15.5 H-GSP 8426 M3 3 0/0 H-GSP 8114 M3 0 0/0 H-GSP 8424 M3 5 14.275 13.5 4/3 13.167 12.9-13.7 H-GSP 8114a M3 2 0/1 12.500 H-GSP 8106 M3 9 14.280 13.8-14.7 5/4 12.700 11.8-13.3 H-GSP 8311 M3 15 14.985 13.8-16.4 13/15 12.460 11.4-13.5 H-GSP 8107 (a) M3 4 14.9 14.4-15.4 2/3 13.000 12.0-14.1 In 0.1 -mm units. 162 Annals of Carnegie Museum VOL. 70 Appendix 2 Length/Width ratios of several Eumyarion and Prokanisamys species. Species LAV M' L/W M2 LAV M, LAV M. Localities source of data Eumyarion medium 40 42 28 33 Sansan (Lartet, 1851) 1.39 1.00 1.59 1.20 Coll. UU Eumyarion latior 819 8 9 7 Aliveri (Schaub and Zapfe, 1953) 1.40 1.08 1.60 1.27 Kleinhofmeijer and de Braijn, 1988 Eumyarion bifidus 10 10 10 10 Sandelzhausen (Fahlbusch, 1964) 1.52 1.10 1.68 1.21 Coll. Munchen Eumyarion bifidus 62157 67 65 53/54 Puttenhausen 1.55 1.11 1.66 1.25 Wu, 1982 Eumyarion montanus 65166 59/58 57/58 65 Kesekoy de Bmijn and Sara?, 1991 1.49 1.09 1.62 1.24 De Braijn and Sarag, 1991 Eumyarion carbonicus 63164 69/68 63/64 75/76 Harami 1 1.61 1.08 1.59 1.20 De Braijn and Sarag, 1991 Eumyarion aff. carbonicus 618 7/6 11/12 9 Harami 3 1.50 1.06 1.58 1.32 De Braijn and Sarag, 1991 Prokanisamys arifi 6 6/4 6 11 H-GSP 116 de Braijn et al., 1981 1.39 1.0 1.49 1.15 De Braijn et al., 1982 Prokanisamys arifi 516 10 10/9 13/14 H-GSP 8311 1.32 1.05 1.47 1.11 This article Prokanisamys major n. sp. 2 1 2 5/4 H-GSP 8224 1.31 1.07 1.41 1.12 This article Prokanisamys major 2 1 1 H-GSP 8114 1.30 1.03 1.44 This article Prokanisamys benjavuni 19 14/13 13/14 16/15 Li (Mein and Ginsburg, 1985) 1.21 1.05 1.49 1.20 Mein and Ginsburg, 1997 Prokanisamys benjavuni 5 3 3/5 5/4 H-GSP 8114 1.21 0.95 1.48 1.00 This article Prokanisamys benjavuni 312 2/3 2 3 H-GSP 8424 1.26 1.08 1.40 1.09 This article Prokanisamys kowalskii 9112 10/12 5/6 5/4 z 113 (Lindsay, 1996) 1.40 1.01 1.40 1.11 Lindsay, 1996 2001 Wessels and de Bruun — Miocene Rhizomyidae from Pakistan 163 Appendix 3 Material and measurements of Prokanisamys major nov. sp. Prokanisamys major Locality Length Width N Mean Range N/N Mean Range H-GSP 8214 M> 1 0/0 H~GSP 8224 Ml 2 22.200 1/1 17.000 H-GSP 8427 Ml 0 H-GSP 8227 Ml 0 H-GSP 8425 Ml 0 H-GSP 8114 Ml 2 21.000 21.0-21.0 2/2 16.050 15.9-16.2 H-GSP 8106 Ml 1 0/1 15.7 H-GSP 8214 M2 0 H-GSP 8224 M2 1 18.000 1/1 16.900 H-GSP 8427 M2 0 H-GSP 8227 M2 2 16.900 1/1 16.900 H-GSP 8425 M2 0 H-GSP 8114 M2 1 16.100 1/1 15.600 H-GSP 8106 M2 2 16.150 16.0^16.3 2/2 15.150 14.3-16.0 H-GSP 8214 M3 2 15.700 1/1 15.400 H-GSP 8224 M3 7 15.275 14.5-16.0 4/4 16.300 16.0-16.6 H-GSP 8427 M3 1 16.600 1/1 17.300 H-GSP 8227 M3 4 15.500 15.2-15.8 2/2 16.000 15.6-16.4 H-GSP 8425 M3 0 H-GSP 8114 M3 5 15.350 14.6-16.1 2/2 17.100 16.0-18.2 H-GSP 8106 M3 0 H-GSP 8214 M, 1 19.500 1/1 14.600 H-GSP 8224 Ml 4 21.500 21.2-21.8 2/2 15.033 14.1-16.1 H-GSP 8427 Ml 0 H-GSP 8227 Ml 3 19.267 19.0-19.4 3/3 14.200 13.7-14.9 H-GSP 8425 Ml 1 20.600 1/1 14.500 H-GSP 8114 Ml 4 19.400 1/1 13.500 H-GSP 8106 Ml 0 H-GSP 8214 M2 1 18.500 1/1 16.800 H-GSP 8224 M2 9 19.640 17.5-21.3 5/4 17.525 15.6-18.8 H-GSP 8427 M2 1 18.500 1/1 16.800 H-GSP 8227 M2 1 20.200 1/1 17.000 H-GSP 8425 M2 0 H-GSP 8114 M2 1 0/0 H-GSP 8106 M2 1 17.7 1/1 16.2 H-GSP 8214 M3 0 H-GSP 8224 M3 1 16.300 1/1 16.300 H-GSP 8427 M3 1 0/0 H-GSP 8227 M3 1 0/0 H-GSP 8425 M3 0 H-GSP 8114 M3 9 15.057 13.4-16.4 7/6 13.300 12.7-14.1 H-GSP 8106 M3 2 16.800 1/1 13.800 In 0.1 -mm units. 164 Annals of Carnegie Museum VOL. 70 Appendix 4 Material and measurements o/ Prokanisamys benjavuni. Prokanisamys benjavuni Locality Length Width N Mean Range N/N Mean Range H-GSP 8426 M' 1 0/1 14.800 H-GSP 8114 M‘ 5 19.580 18.1-21.2 5/5 16.140 15.2-17.6 H-GSP 8424 M' 7 18.600 17.8-19.7 3/2 14.800 14.6-15.0 H-GSP 8114a M' 4 19.650 19.0-20.4 4/4 16.250 14.3-18.1 H-GSP 8426 M2 2 17.950 17.9-18.0 2/2 16.900 13.6-17.2 H-GSP 8114 M2 3 15.133 14.6-15.4 3/3 15.933 15.0-16.5 H-GSP 8424 M2 3 16.900 16.6-17.2 2/3 15.633 15.0-16.2 H-GSP 8114a M2 5 16.375 15.3-17.7 4/4 17.525 16.3-18.4 H-GSP 8426 M3 2 17.000 1/1 16.700 H-GSP 8114 M3 8 16.383 15.7-17.1 6/6 16.083 15.0-17.7 H-GSP 8424 M3 2 17.800 1/1 17.7 H-GSP 8114a M3 1 16.600 1/1 16.000 H-GSP 8426 M. 1 19.400 1/1 12.500 H-GSP 8114 M, 6 19.467 19.3-19.6 3/5 13.200 12.1-14.0 H-GSP 8424 M, 5 18.600 18.5-18.7 2/2 13.250 12.5-14.0 H-GSP 8114a M, 1 0/1 12.900 H-GSP 8426 M^ 3 16.333 16.0-16.6 3/3 15.167 14.1-16.1 H-GSP 8114 M^ 6 16.700 16.2-17.6 5/4 16.625 15.6-17.5 H-GSP 8424 M^ 6 17.300 17.1-17.5 3/3 15.933 14.4-17.1 H-GSP 8114a M^ 6 18.060 16.5-19.7 5/5 17.300 16.2-18.4 H-GSP 8426 M3 7 19.067 17.6-20.1 3/5 15.900 14.7-17.1 H-GSP 8114 M3 5 18.350 17.6-18.7 4/4 15.275 14.1-16.3 H-GSP 8424 M3 7 19.133 16.5-21.2 3/4 15.300 14.1-16.9 H-GSP 8114a M3 7 19.150 17.6-21.3 4/4 15.975 15.3-16.5 In 0.1-mm units. Appendix 5 Material and measurements of .^Prokanisamys sp. A. IProkanisamys sp. A Length Width Locality N Mean Range N/N Mean Range H-GSP 8107 (a) M’ 1 0/0 M2 0 M3 0 M, 1 20.800 1/1 14.100 M3 1 18.200 1/0 M3 2 19.800 1/2 16.800 16.3-17.3 In 0.1-mm units. 2001 Wessels and de Bruijn — Miocene Rhizomyidae from Pakistan 165 Appendix 6 Material and measurements o/ Kanisamys indicus. Kanisamys indicus Locality Length Width N Mean Range N/N Mean Range H-GSP 8214 M' 0 H-GSP 8224 M* 3 23.867 22.9-24.9 3/3 19.233 18.3-20.1 H-GSP 8427 M' 3 23.000 1/3 18.250 18.2-18.3 H-GSP 8227 M> 2 24.800 1/2 18.850 17.9-19.8 H-GSP 8425 M' 0 H-GSP 8114 M' 5 23.767 22.0-24.8 3/4 17.950 17.0-18.7 H-GSP 8214 M2 2 19.200 19.1-19.3 2/2 19.750 19.1-20.4 H-GSP 8224 M2 8 18.440 17.9-19.8 5/6 19.250 18.3-19.8 H-GSP 8427 M2 0 H-GSP 8227 M2 2 21.200 1/2 19.750 19.0-20.5 H-GSP 8425 M2 2 19.400 1/1 20.800 H-GSP 8114 M2 1 20.000 1/1 19.600 H-GSP 8214 M3 1 16.700 1/1 18.600 H-GSP 8224 M3 5 17.925 17.2-19.2 4/4 19.500 17.8-21.8 H-GSP 8427 M3 0 H-GSP 8227 M3 1 18.500 1/1 19.200 H-GSP 8425 M3 1 0/0 H-GSP 8114 M3 1 0/0 H-GSP 8214 M, 0 H-GSP 8224 M, 5 22.775 22.3-23.3 4/3 17.733 16.8-18.4 H-GSP 8427 M, 0 H-GSP 8227 M, 1 22.100 1/1 16.700 H-GSP 8425 M, 0 H-GSP 8114 M. 0 H-GSP 8214 M^ 0 H-GSP 8224 M^ 4 22.400 21.2-23.6 2/2 21.500 21.0-22.0 H-GSP 8427 M^ 0 H-GSP 8227 4 21.000 19.6-21.7 3/4 19.750 19.0-20.2 H-GSP 8425 M2 1 0/1 19.800 H-GSP 8114 M2 0 H-GSP 8214 M3 0 H-GSP 8224 M3 9 22.183 20.8-24.2 6/6 19.483 18.2-21.8 H-GSP 8427 M3 0 H-GSP 8227 M3 9 21.400 20.4-22.2 3/5 18.280 16.9-18.8 H-GSP 8425 M3 2 0/0 H-GSP 8114 M3 4 21.333 19.3-22.6 3/1 18.000 In 0.1 -mm units. 166 Annals of Carnegie Museum VOL, 70 Appendix 7 MjIM^ratio, Width M./Mj P. arifi P. major P. benjavuni K. indicus H-GSP 8214 H-GSP 8224 H-GSP 8427 0.92 0.91 H-GSP 8227 H=GSP 8426 H-GSP 8425 0.79 0.91 H-GSP 8114 1.12 0.86 H-GSP 8424 0.84 0.87 H-GSP 8114a 1.01 0.81 H-GSP 8106 1.00 H-GSP 8311 0.96 H-GSP 8107 (a) 0.97 MURREE Prokanisamys arifi 1,03. THAILAND Prokanisamys benjavuni 0.84. ZINDA PIR DOME Prokanisamys kowalskii 0.94. POTWAR PLATEAU Prokanisamys major (specimen 26046, loc 640) 0.93. Appendix 8 Area occlusal surface MfArea occlusal surface X 100. Area M,/Area M3 X 100 P. arifi P. major P. benjavuni IP. sp.A K. indicus H-GSP 8214 H-GSP 8224 H-GSP 8427 121 93 H-GSP 8227 H-GSP 8426 H-GSP 8425 80 94 H-GSP 8114 130 91 H-GSP 8424 H-GSP 8114a 101 85 H-GSP 8106 128 H-GSP 8311 111 H-GSP 8107 (a) 121 88 MURREE Prokanisamys arifi 127. THAILAND Prokanisamys benjavuni 86. ZINDA PIR DOME Prokanisamys kowalskii 111. POTWAR PLATEAU (specimen 26046, loc 640) Prokanisamys major 104. 2001 Wessels and de Bruijn — Miocene Rhizomyidae from Pakistan 167 Appendix 9 Area occlusal surface M^. Area In 0.1 -mm units P. arifi P. major P. benjavuni K. indicus H-GSP 8214 379.20 H--GSP 8224 304.2 354.97 H-GSP 8427 H-GSP 8227 285.61 418.70 H-GSP 8426 304.20 H=GSP 8425 403.52 H^GSP 8114 251.16 241.60 392.00 H--GSP 8424 263.64 H^GSP 8114a 217.54 287.00 H=GSP 8106 231.81 246.24 H-GSP 8311 211.58 H--GSP 8107 (a) 218.95 MURREE Prokanisamys arifi 204.49. THAILAND Prokanisamys benjavuni 298.8. ZINDA PIR DOME Prokanisamys kowalskii 282.23. POTWAR PLATEAU Prokanisamys major (specimen 26042) 264. Appendix 10 Area occlusal surface M2. Area M2 In 0.1 -mm units P. arifi P. major P. benjavuni K. indicus H^GSP 8214 310.80 H-GSP 8224 343.00 481.60 H-GSP 8427 310.80 H-GSP 8227 343.40 414.75 H^GSP 8426 247.76 H==GSP 8425 H-GSP 8114 277.22 H=GSP 8424 275.07 H-GSP 8114a 252.17 313.30 H--GSP 8106 221.20 286.74 H^GSP 8311 222.78 H^GSP 8107 (a) 237.80 MURREE Prokanisamys arifi 213.52. THAILAND Prokanisamys benjavuni 266.71. ZINDA PIR DOME Prokanisamys kowalskii 286.58. POTWAR PLATEAU Prokanisamys major (specimen 26046) 351. 168 Annals of Carnegie Museum VOL. 70 Appendix 11 Prokanisamys arifi Maximum of measured crown height of protocone(id)/hypocone(id) Murree H-GSP 8107 H-GSP 8311 H-GSP 8106 H-GSP 8114a H-GSP 8114 H-GSP 8424 H-GSP 8426 M' 10.3/11.5 9.5/9.5 11.1/12.4 10.0/11.1 7.4/8.0 8.8/8.8 M2 9.5/9.0 10.0/10.8 10.7/10.0 11.5/10.3 16.0/13.6 M3 9.0 9.0 9.9 8.7 8.4 10.8 12.5 11.6 M, 7.4/9.8 8. 8/9.2 8. 0/8. 6 7.4/9.8 10.2/8.7 7. 9/7. 9 5. 9/5. 9 M^ 8.7/8.9 9.5/9.5 10.7/10.5 9.2/8.2 9.5/10.1 7.0 M3 8.5/8.0 9.5/9.0 8.2/7.8 9.2/8.2 12.5 12.4 In 0.1 -mm units Prokanisamys major Maximum of measured crown height of protocone(id)/hypocone(id) H-GSP 8106 H-GSP 8114 H-GSP 8425 H-GSP 8227 H-GSP 8224 H-GSP 8214 M' 10.5/8.8 14.9/14.5 M2 10.7/10.3 9.6/— 13.6/12.4 M3 9.2 7.9 11.0 14.0 M, 7.8/9.3 8.3/— 9.21— 9.0/10.5 M, 10.4/9.2 10.5/— 8.3/— 8.2/— 9.1/10.5 M3 9.1 9.0 10.0 In 0.1 -mm units Prokanisamys benjavuni Maximum of measured crown height of protocone(id)/hypocone(id) H-GSP 8114a H-GSP 8424 H-GSP 8114 H-GSP 8426 M' 14.0/16.6 14.3/14.3 12.4/13.1 — /16.6 M2 6.6/15.3 17.4/16.8 14.0/13.3 15.4/14.7 M3 15.7 9.5 13.4 13.0 M, 9.1/11.8 12.2/8.0 11.5/11.5 7.8/10.5 M2 13.5/15.2 10.6/— 11.6/13.5 14.0/11.0 M3 11.6 13.2/13.0 10.7/9.6 In 0.1 -mm units Kanisamys indicus Maximum of measured crown height of protocone(id)/hypocone(id) H-GSP 8114 H-GSP 8425 H-GSP 8227 H-GSP 8427 H-GSP 8224 H-GSP 8214 M* 13.2/15.6 13.3/14.3 14.1/13.3 14.5/15.4 M2 11.4/9.6 15.0/15.0 14.4/15.5 14.0/14.0 M3 12.5 20.6 4.1 M, 10.5/12.6 14.5/15.4 M2 8.3/— 12.3/12.3 9.6 M3 13.8/15.0 7.0 13.6/13.4 16.0/17.8 In 0.1 -mm units ? Prokanisamys sp. A Maximum of measured crown height of protocone(id)/hypocone(id) H-GSP 8107 (a) M‘ 9.5/9.5 M2 M3 M, 16.8/19.8 M2 M3 8.3/9.2 In 0.1 -mm units ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 2, Pp. 169-178 24 May 2001 FLEAS (SIPHONAPTERA: CTENOPHTHALMIDAE AND RHOPALOPSYLLIDAE) FROM ARGENTINA AND CHILE WITH TWO NEW SPECIES FROM THE ROCK RAT, ACONAEMYS FUSCUS, IN CHILE Michael W. Hastriter’ Research Associate, Section of Invertebrate Zoology Abstract Two new species of fleas (Rhopalopsyllidae) collected from the Rock Rat, Aconaemys fuscus (Wa- terhouse, 1842), are described from the Provinces of Nuble and Talca, Chile. They are Ectinorus gailardoi n.sp. and Ectinorus mondacai n. sp. Distribution records for other species from Chile and Argentina are also included. The current number of species of fleas known to occur in Chile is 94. Key Words: Ectinorus gailardoi, Ectinorus mondacai, Rhopalopsyllidae, flea, rock rat Introduction Fleas were collected by Milton H. Gallardo and Fredy Mondaca, Instituto de Ecologia y Evolution, Universidad Austral de Chile, Valdivia, Chile, during phy- logenetic studies of the rodent family Octodontidae in Argentina and Chile. Among these collections were two new species of Ectinorus Jordan, 1942 and some other flea records. The genus Ectinorus, represented by 29 species, is confined to montane regions of Peru, Chile, Bolivia and Argentina, occurring primarily on octodontid rodents. Jordan (1942) erected the genera Dysmicus and Ectinorus, a systematic scheme that was followed by Macchiavello (1948) and Johnson (1957). Smit (1968) rel- egated the genus Dysmicus to a subgenus of Ectinorus, whereas Lewis (1976) further defined the subgeneric taxa Dysmicus and Ectinorus (exclusive of a single species assigned to the subgenus Panallius Jordan, 1942). Lewis (1976) also grouped the subspecies of E. onychius within the subgenus Ectinorus. Smit (1987) provides the latest comprehensive treatment of Ectinorus, recognizing three sub- genera [Ectinorus (22 species), Ichyonus (four subspecies), and Panallius (one species)]. A subsequent description of two additional species [E. (Ectinorus) cur- vatus and E. (Ectinorus) uncinatus] provided by Beaucournu and Gallardo (1991) totals 29 species of Ectinorus. Two additional descriptions of species of Ectinorus follow. Materials and Methods Measurements of tibial and tarsal segments (tarsal claws and petioles excluded) of holotypes and allotypes were recorded in microns, and the first entries for each appendage are those of the holotype followed by the allotype (reference Tables 1 and 2). The overall body dimensions of males and females were measured from the foremost portion of the frons to the apex of the st. VIII in males and to the posterior border of the sensilial plate in females. Illustrations were prepared with 'Monte L. Bean Life Science Museum, Brigham Young University, 290 MLBM, RO. Box 20200, Provo, Utah 84602-0200, e-mail: hastritermw@sprintmail.com. Submitted 22 November 2000. 169 170 Annals of Carnegie Museum VOL. 70 Table 1 . — Measurements of tibial and tarsal segments of E. gallardoi n. sp. Segments Tibia Tarsus I Tarsus II Tarsus III Tarsus IV Tarsus V Fore leg 96 37 37 37 35 62 143 49 49 47 40 86 Mid leg 193 64 84 64 35 74 217 77 116 84 42 111 Hind leg 272 153 116 77 35 89 344 210 128 89 49 114 the aid of a camera lucida mounted on a Carl Zeiss® compound microscope. Terminology of flea structures follows that of Rothschild and Traub (1971). Systematic Entomology Ctenophthalmidae Neotyphloceras crassispina crassispina Rothschild, 1914 Specimens Examined. — CHILE. Valparaiso Prov.: Parque Nacional Lago Penuelas (33°08’S, 71°3rW), ex Octodon degus (Molina, 1782), 14 October 1998, E Mondaca, Id. Neotyphloceras crassispina chilensis Jordan, 1936 Specimens Examined. — CHILE. Valparaiso Prov.^ Parque Nacional Lago Penuelas (33°08’S, 71°3rW), ex Octodon lunatus Osgood, 1943, 4 September 1999, E Mondaca, IS, 7$; ex Abrocoma bennetti Waterhouse, 1837, 14 October 1998, F. Mondaca, Id. Neotyphloceras crassispina ssp.? Specimens Examined. — CHILE. Valparaiso Prov.j Parque Nacional Lago Penuelas (33°08’S, 71°3LW), ex Thylamys elegans (Waterhouse, 1839), 8 January 1998, E Mondaca, 2$. Rhopalopsyllidae Ectinorus (Ectinorus) gallardoi, new species (Fig. 1A-=F) Type Material— Cmi.'E. Nuble Prov.: Chilian Springs (36°54’S, 71°3rW), ex: A. fuscus, 29 February 1998, M. H. Gallardo, Id paratype. Talca Prov.: Reserva Nacional Radal 7 Tazas, (35°28’S, 70°59’W), ex: Rock Rat, Aconaemys fuscus (Waterhouse, 1842), 24 March 1999, Fredy Mondaca, holotype, allotype and 2d, 1$ paratypes. Reserva Nacional Altos de Lircay (35°36’S, 71°12’W), ex: Table 2. — Measurements of tibial and tarsal segments of E. mondacai n. sp. Segments Tibia Tarsus I Tarsus II Tarsus III Tarsus IV Tarsus V Fore leg 165 49 52 44 40 99 153 57 64 49 42 111 Mid leg 254 96 111 96 42 109 262 104 121 67 44 114 Hind leg 363 222 151 91 52 116 387 242 161 94 47 121 2001 Hastriter — ^Two New Fleas from Chilean Rock Rats 171 E F Fig. 1. — Ectinorus gallardoi n. sp. A. Head and thorax, male holotype. B. Male clasper (basimere and telomere). C. Eighth and ninth stemites, male, D. Aedeagus. E. Female seventh sternum. F. Sperma- theca. Scale =1 00 jx. 172 Annals of Carnegie Museum VOL. 70 A. fuscus, 28 February 1998, M. H. Gallardo, 17 d, 8$ paratypes. Holotype (USNM 105736), allotype and four paratypes {26 , 2$) deposited in the National Museum of Natural History, Washington, D. C.; six paratypes (3(3, 3$) in the Carnegie Museum of Natural History, Pittsburgh, PA (Accession No. 36,734); seven paratypes (43, 3 9) in the Museo Nacional Historia Natural (Zoology Section), San= tiago, Chile; remaining paratypes (113, 19) in the collection of the author. Diagnosis. — Males and females of this new species key to the couplet of E. cocyti and E. chilensis according to Smit (1987). Males may be separated from those of E. cocyti by the absence of a conspicuous dorsal spur arising from the dorsal extension of the lamina of the aedeagus near the base of the median dorsal lobe and from both E. cocyti and E. chilensis by the presence of an apically pointed median dorsal lobe, an apically pointed crochet, the lobe on the dorsal caudal margin of st. VIII is more pronounced, and other details of the aedeagus. Females are not separable from E. cocyti and E. chilensis, although host associ- ations may be a distinguishing factor. Ectinorus gallardoi appears to prefer the single host A. fuscus (note collection records from three distinct localities), where- as E. cocyti and E. chilensis have never been reported from this host. This new species is also separable from the newly described species below {E. mondacai) by the absence of overlapping lobes on the posterior margin of the st. VIII. Description. — Head (Fig. lA); Frons somewhat bluntly rounded in male, less so in female. Stout upturned tubercle dorsad to oral angle. Ocular row of 4 setae with intercalaries. Occipital area with three rows of a single seta each and main row with 5 setae on one side and 6 on other with additional intercalaries. Pigmented eye well developed with ventral sinus, 1-2 minute setae behind eye on genal lobe and a single stout seta just below eye (not marginal). Caudal margin of genal lobe slightly concave. Setae of scape and pedicel at most reaching the next segment. Clavus extending onto pro- pleurum in both sexes, especially the male. Shallow occipital sulcus present in male. Labial palpus of five segments extending to trochanter. Thorax (Fig. lA): Pronotum with two rows of setae (anterior row: 3 per side male, 6 per side female; main row: 7 per side). Propleuron with dorsal depression to accommodate apex of antenna. Mesonotum with two rows of setae in male (anterior row: 5-6 per side; main row: 5 per side) and three in female (anterior row: 2-3 minute setae per side; middle and main row: each 6 per side). Flange of mesonotum with 8-9 pseudosetae per side in male and 10-11 per side in female. Mesepisternum 1 seta, mesepimeron 4 setae. Mesosternum reduced to a sclerotized point. Metanotum with two rows of setae in male (anterior row: 5; main row: 5) and three rows in female (anterior row: 2-3 minute setae per side; middle row: 8 per side; main row: 5 per side). Lateral metanotal area with 2 setae, pleural arch well developed. Metepisternum with 1 seta, metasternum anteriorly rounded and not protruding downward. Metepimeron with two vertical rows of setae variable from 2-5 in each row, but usually 3 in each row. Legs (Table 1): Procoxa has 20—21 lateral setae, none of which is marginal. Anterior margins of meso- and metacoxae with setae from base to apex on lateral surfaces. Fore femur with lateral group of 4-5 setae and single mesal seta, mid and hind femora each with ventral row of 5-6 lateral setae and 2-3 on mesal surface. The lateral femoral-tibial bristle longer of two on fore femur and shorter on mid and hind femora. Dorsal margin of fore, mid and hind tibia with 6, 7, and 7 notches, respectively. A row of 7-8 setae adorns the lateral surface of mid and hind tibiae. Four lateral plantar bristles on all tarsal segments, the apical pair distinctly separated from the proximal three pairs, which are evenly spaced. Unmodified Abdominal Segments: Males with two rows of setae on t. I (anterior row: 5 per side; main row: 6 per side) and a single row on t. II-VII (8 per side), one below level of rounded spiracles. Females have two rows per side on all segments of t. I- VII (anterior row: 5, 7, 6, 7, 5, 5, 5 per side, respectively; and main row: 6, 8, 8, 8, 8, 8, 5 per side, respectively). Marginal spinelets on t. I vary from 3-4 per side. Two setae are below level of spiracle on t. III-VI of female. Single antesensilial bristle recessed in notch in female and marginal in male. Males have single row of setae on st. II-VII (1, 3, 3, 3, 3, 3 per side, respectively) and females have single minute seta per side on st. II, two rows on st. III-V (anterior row: each 1 per side; and main row: each 4 per side), and single row of 5 per side on st. VI. Modified Abdominal Segments, male (Fig. IB-C): Tergum VIII inconspicuous, Basimere of t. IX entire along apicoventral margin. Telomere small, pointed apically, with ventral portion extended far below level of articulation. Acetabular bristle small. Manubrium spatulate and broadest apically. Margin of st. VIII entire with broadly rounded dorsocaudal lobe, lateral row of 3-4 small setae. Sternum IX rod-like with row of fine evenly dispersed setae along ventral margin. Tendon of st. IX absent. Aedeagus (Fig ID): Crochet notably sclerotized, pointed apically, bearing ventral membranous lobe. Sclerotized inner tube thick 2001 Hastriter — Two New Fleas from Chilean Rock Rats 173 walled, adorned with heavy thorn-like process on the dorsal margin. Paired median dorsal lobe deli- cately membranous with bifid, heavily sclerotized apical sclerites, both enveloped within faintly visible membranous lateral lobes. The apical sclerites of median dorsal lobe are “hinged” basally to a scler- otized extension of the median lamina, which forms the dorsal portion of the end chamber. Though not coiled, the penis rods extend far beyond the apex of the aedeagal apodeme. Modified Abdominal Segments, female (Fig. lE-F): Tergum VIII bearing row of 3-4 small setae anterior to a vermiform spiracle. Laterally t. VIII has two rows of setae (7-8 small and 2-3 stout setae per side), 8-9 marginals, and a group of 8-10 setae on a lobe beneath the apex of the sclerite. The anal stylet arises from the ventrolateral margin of the dorsal anal lobe and bears a single apical bristle. Sternum VII is without sinuses or lobes, although the caudal margin is slightly undulate and is adorned with a main row of 7 setae and 2-3 small setae anterior to these. Bursa copulatrix is thin, sclerotized and apically reflected caudad. The bulga of the spermatheca is slightly longer than wide, the hilla exceeding the length of the bulga. Dimensions (slide mounted specimens). — Male, 1.7 mm (n=21), female, 2.2 mm (n=10). Etymology. — Dr. Milton H. Gallardo, Instituto de Ecologia y Evolution, Universidad Austral de Chile, Valdivia, Chile, has contributed significantly to an understanding of the phylogeny of octodontid rodents in the Southern Cone of South America. During these studies, he devoted considerable time to collecting ectoparasites. This new species is thus named in his honor. Remarks. — All specimens were collected from A. fuscus at several sites sepa- rated by several hundred kilometers. Since it occurs in the central range of A. fuscus, it may have a wider distribution to the north and to the south of current records. Ectinorus (Ectinorus) mondacai, new species (Fig. 2A~N) Type Material. — CHILE. Nuble Prov.: Chilian Springs (36°54’S, 71°3rW), ex: A. fuscus, 29 February 1998, M. H. Gallardo, allotype and 26 , 69 paratypes. Talca Prov.: Radal 7 Tazas Nacional Parque (35°28’S, 70°59’W), ex A. fuscus, 24 February 1999, Fredy Mondaca, holotype and 26, 2$ paratypes. Holotype (USNM 105735), allotype, and two paratypes (16, 19) deposited in the National Museum of Natural History, Washington, D. C.; four paratypes (16, 3 9) in the Museo Nacional Historia Natural (Zoology Section), Santiago, Chile, remaining paratypes (26, 49) in the collection of the author. Diagnosis. — Most closely related to E. martini to which both males and females key according to Smit (1987). Males may be distinguished from E. martini by the shape of the st. IX which is mitten shaped, presence of a deep cleft in the lateral lobe of aedeagus, median dorsal lobe extremely sclerotized, sclerotized inner tube with dorsal and ventral lobes, and ventral aedeagal pouch ending about half the distance of the length of the median lamina; whereas in E. martini, the St. IX is not mitten shaped, the cleft in the lateral lobe is shallow, median dorsal lobe less sclerotized, sclerotized inner tube is without lobes, and ventral pouch of aedeagus extends to near apex of median lamina. Females of E. martini and E. mondacai are indistinguishable. Description. — Head (Fig 2A): Frons evenly rounded with 2 marginal placoid pits evenly spaced between a well-developed frontal tubercle and the antennal fossa. Preantennal area with an uneven ocular row of 2 medium and 2 larger setae with several scattered small setae at level of eye in male and 3 setae in ocular row of female. Eye medium size with ventral sinus, darkly pigmented and confluent with heavily sclerotized margins of tentorium. Anterior arm of tentorium clearly visible anterior to eye. Single heavy seta at posteroventral margin of eye and 2 minute setae on apical lobe that is subtended by shallow sinus. Shallow occipital groove extends from falx to posterior margin of head in male. Occipital region with three placoid pits, four rows of setae (single seta in each of three anterior rows and 4-5 with intercalaries in posterior row) in male and two rows in female (single seta 174 Annals of Carnegie Museum VOL. 70 Fig. 2. — Ectinorus mondacai n. sp. A. Head and thorax, male holotype. B. Male clasper. C-G. Lateral apical aspect of the distal arm of male ninth sternum (paired) of five specimens (L^left, R= right per pair). H. Aedeagus. I. Female seventh sternum. J. Spermatheca. K-N. Ectinorus levipes, lateral apical aspect of the distal arm of male ninth sternum (paired) of four specimens (L=left, R=right per pair). Scale= lOOp.. 2001 Hastriter — Two New Fleas from Chilean Rock Rats 175 in anterior row and 5~6 with intercalaries in posterior row). Antennal groove with dense row of minute setae along dorsal margin (only two minute setae in female). Antenna extending onto prosternum. Antennal scape of male with minute setae at base and fringe of short setae along apical flange; flange of pedicel with several minute setae along dorsal margin and a single seta at ventral margin. Pedicel of female with interior lobe bearing 1 small seta and a single stout seta that extends half the length of the clavus. Epipharynx, maxillary lacinia and 5 segmented labial palp extending to apex of fore coxa. Maxilla acutely pointed; anterior margin convex. Four segmented maxillary palp extending ca. two-thirds the length of fore coxa. Thorax (Fig. 2A): Pronotum higher than long, bearing single row of 6-7 setae per side with intercalaries. Prosternosome broadening posterad. Mesonotum with anterior row of 3 setae and posterior row of 5 larger setae per side with intercalaries. Mesonotal collar with 9-10 pseudosetae per side whose lengths extend well beyond margin of collar. Mesepimeron bearing 3 setae; mesepisternum with single seta. Metanotum with 3 rows of setae in male; per side, single small dorsal in anterior row, 3-4 in middle row and 7 in posterior row with intercalaries. Ventral seta in main row distinctly separated from those dorsad. Female with two rows, each with 5 setae per side. Lateral metanotal area with 2 setae, 1 long and 1 short. Pleural arch well developed with very robust pleural ridge. Metepisternum with single seta. Metepimeron with distinct broad lobe along caudal margin; two rows of setae, anterior row with 3 setae (upper seta dorsad to spiracle), posterior with 3 (upper seta ventrad to spiracle). Female has 2 setae in anterior row and 3 on posterior, all ventrad to spiracle. Legs (Table 2): Procoxa with 28-30 lateral setae (including marginals). Fore femur with 11- 13 lateral setae, 2 mesal setae, 2 large postero ventral marginal setae, fringe of minute setae along dorsal margin and 2 large bristles guarding femoral-tibial joint (mesal bristle smaller of two). Tibia with 5 dorsal notches (2, 2, 3, 2 and 3 bristles, proximal to distal) and a vertical lateral row of 4 small setae. Outer internal ridge of mesocoxa only developed dorsally. Suture dividing outer surface of mesocoxa at most indicated. Mid femur with 8-9 lateral setae, mesal row of 5 setae (3 in female), marginal row along dorsum, single seta at postero-ventral margin, and 2 large bristles guarding fem- oral-tibial joint (mesal larger of two). Tibia bearing 6 dorsal notches (2, 2, 2, 3, 2, 3 bristles, proximal to distal) and lateral row of 5 setae. Metacoxa with well-developed and complete outer internal ridge; patch of slender setae extending along anterior lateral margin increasing in number from base to apex. Hind femur with two lateral rows of setae (7 ventrad, 1 dorsad) (one lateral row in female of 6 setae), ventral mesal row of 9 setae (7 in female), dorsal margin with fringe of setae, and a single postero- ventral seta. Arrangement of two large bristles guarding femoral-tibial joint is the same as for that of mid femur. Sculpturing of lateral cuticular surface densely striated and mesal surface coarsely sculp- tured perpendicular to long axis of appendage. Tibia with 6 dorsal notches (2, 2, 2, 3, 2, 3 bristles, proximal to distal), two lateral rows of setae (ventral row of 9 and row of 9 adjacent to dorsal notches), and prominent apical tooth. Mesal surface bears minute tubercles along coarse reticulations. Disto- tarsomeres of all segments bear 4 pairs lateral plantar bristles, 2 pre-apical plantar bristles (semi- spiniform — 1 long, 1 short), and pair of pre-apical plantar hairs. Unmodified Abdominal Segments: Abdominal segments (tergites and stemites) appear banded because of dark sclerotizations cephalad and lighter cuticle caudad. Two rows of setae per tergite, all main rows bearing intercalary setae. Considerable variation exists in number of setae from specimen to specimen among males and females. Setae per side in main row on t. I- VII; 5-6, 8-10, 8-9, 7-9, 7-9, 7-8, 6-7, in males and 4-7, 8-10, 8-10, 7-9, 7-8, 6-8, 4-6, in females, respectively. Marginal spinelets per side on t. I: 2-4 in males, 0-4 in females. Single antesensilial bristle borne on a peduncle recessed in a V-shaped notch. Setae per side in main row on st. II- VII; 2-3, 3, 3, 3, 3, 3, in males and 2-3, 4, 3, 3, 3, 4, in females. Stemite II of males and females with lateral patch of 5-6 and 13-14 small setae on anterodorsal surface, respectively. Modified Abdominal Segments, male (Figs. 2B, 2C-G): Tergite IX similar to that of E. martini. Basimere with narrow, strongly upturned manubrium, body apically extended into a bluntly acute lobe subtended by a shallow sinus and broad ventral lobe. Ventral lobe fringed with long evenly spaced setae. Telomere slightly concave along ventral margin, apex bluntly square and extending beyond apical lobe of basimere. Sternum VIII distinctly more heavily sclerotized than preceding segments; apex obliquely blunt with extreme sclerotization, 2-3 lateral setae subtended by 5-6 smaller setae. Distal arm of st. IX resembling a mitten with thumb extended; bearing a fringe of setae along ventral margin. The apical mitten-like clubs of the paired distal arms of st. IX are asym- metrical on all five specimens; the apex of the thumb of the mitten on the left side extends to the apex of the mitten’s fingers, whereas the apex of the fingers are extended far beyond the thumb on the right side, Aedeagus (Fig. 2H): Dorsal median lobe heavily sclerotized, lobe nearly right angled anterodorsally, terminating in a rounded point. Lateral lobes bifurcate into two lobes, dorsal lobe acute at apex and ventral lobe truncate. Ventral floor of end pouch terminates midway along medial lamina as a sharp projection. Modified Abdominal Segments, female allotype (Figs. 2I-J): Tergum VIII with group of 3-4 small setae dorsal to vermiform spiracle, subtended by vertical row of 5 setae; caudal margin convex, slightly undulating with marginal row of 6 setae. Mesal lobe adorned with 17-18 176 Annals of Carnegie Museum VOL. 70 setae (8 long, 9-10 short). Sensilium with narrow sensilial plate. Anal stylet 2.5 times as long as greatest width, positioned medioventrad on dorsal anal lobe, bearing 2 apical setae (1 long, 1 minute). Ventral anal lobe flattened on dorsal margin; angled ventrally. Caudal margin of st. VII entire with shallow sinus ventrocaudally; bearing row of 4 lateral setae per side. Sternum VIII terminating in a broad, blunt lobe bearing 4-5 minute setae whose alveoli are more conspicuous than setae; lateral surface markedly reticulated. Bulga of spermatheca slightly oval with walls thickened. Terminal half of hilla sclerotized and wider than hyaline proximal portion. Bursa copulatrix with several polymorphic sclerites, dorsum of perula sclerotized. Venter of oviduct sclerotized just anterior to bursa copulatrix. Duct of spermatheca on right side, blind duct on left. Dimensions (slide mounted specimens). — Male, 2.0 mm (n=5), female, 2,5 mm (n=9). Etymology. — The species is named E. mondacai in behalf of Fredy Mondaca, Instituto de Ecologia y Evolution, Universidad Austral de Chile, Valdivia, Chile, whose collection efforts and fieldwork made the discovery of this new species possible. Remarks. — Sympatric with known specimens of E. gallardoi and found solely on A. fuscus, E. mondacai may also be more widely distributed than our records indicate. The shape of the apex of st. IX of all males examined was markedly asymmetrical. Asymmetry of paired genital structures in the Order Siphonaptera was first illustrated by Smit (1987) in a specimens of E. levipes (Jordan and Rothschild, 1923), although no comment was provided. It is noteworthy that it occurs in E. mondacai and also in E. levipes, both in the subgenus Ectinorus. The asymmetry is illustrated in five males of E. mondacai (Fig. 2C-G) and four males of E. levipes (Fig. 2K-N). The evolution or function of this asymmetry can not be explained, since the correlating copulatory structures of the female (st. VII and VIII) are symmetrical. Such odd and consistent genetic expression of asymmetry occurring in like structures (st. IX) among males of two distinct, but closely related taxa, would support their evolutionary affinity. Ectinorus chilensis Lewis, 1976 Specimens Examined. — CHILE. Valparaiso Prov.: Barque Nacional Lago Penuelas (33°08’S, 71°31’W), ex A. bennetti, 14 October 1998, F. Mondaca, 13; ex O. degus, 30 December 1997, F. Mondaca, 1 S . Ectinorus levipes (Jordan and Rothschild, 1923) Specimens Examined. — ^ARGENTINA. Neuquen (1177 m): ex Akodon olivaceus (Waterhouse, 1837), 7 September 1991, D. Gettinger 43; ex Akodon sp., 8 September 1991, Minzenmayer 1$. Other material belonging to the same series in the R. E. Lewis collection that were not examined by the author include the following: Neuquen, (1177 m): ex Phyllotis sp., 7 September 1991, D. Gettin- ger, 13; ex A. olivaceus, 1 September 1991, Minzenmayer, 13, and Neuquen, (793 m): ex A. oli- vaceus, 25 September 1991, D. Gettinger, 13. Remarks. — Robert E. Lewis has eight specimens of E. levipes (IS, 1 $) from Argentina in his collection, heretofore unreported in the literature. Although this species was not found in our study, its apparent asymmetry, paralleling that of E. mondacai, was of interest (see Remarks under E. mondacai above). Dr. Lewis kindly provided hve specimens (4(3, 1 $) for evaluation of asymmetry and granted permission to document records of all eight specimens. Delostichus coxalis (Rothschild, 1909) Specimens Examined. — CHILE. Valparaiso Prov.: Barque Nacional Lago Penuelas (33°08’S, 71°31’W), ex O. lunatus, 14 October 1998, F. Mondaca, 1$; ex O. degus, 14 October 1998, F. Mondaca, 2$; ex A. bennetti, 23 December 1997, F. Mondaca, 1$. 2001 Hastriter — Two New Fleas from Chilean Rock Rats 177 Delostichus incisus Beaucournu and Torres-Mura, 1987 Specimens Examined. — ARGENTINA. San Juan Prov.: Parque Nacional Ischigualasto, ex Octo- mys mimax Thomas, 1920, 8 April 1997, M. H. Gallardo, 4d, 3$. Tetrapsyllus corfidii (Rothschild, 1904) Specimens Examined. — CHILE. Taka Prov.; Reserva Nacional Altos de Lircay, ex A. fuscus, 28 February 1998, M. H. Gallardo, 49, 2d. Valparaiso Prov.: Parque Nacional Lago Penuelas, ex O. degus, 14 October 1998, F. Mondaca Id, 19; ex A. bennetti, 14 October 1998, F. Mondaca, 29; ex A. bennetti, 23 December 1997, F. Mondaca, Id, 19; ex O. lunatus, 14 October 1998, F Mondaca, Id, 39. Discussion The distribution of the three subspecies of N. crassispina {N. c. hemisus Jordan, 1936, N. c. chilensis, and N. c. crassispina) extends from northern Peru to central Bolivia and south through Chile and western Argentina (Smit, 1968). Sympatry of N. c. chilensis and N. c. crassispina in Valparaiso Province promotes hybrid- ization, making identifications at the sub-specific level uncertain. Based on the length, shape, and setation of the apical portion of the fixed process of the clasper [originally proposed as diagnostic criteria by Jordan (1936)], only a single male was clearly N. c. crassispina. The remaining specimens were N. c. chilensis. The sole male N. c. crassispina has a delicate dorsal membranous lobe arising from the dorsal portion of the aedeagus, which is enveloped by the two lobes of the fixed process of the clasper. This membranous lobe is covered with translusent spicules. This structure is also present in all specimens of N. c. crassispina re- cently collected by the author in Ancash Department, Peru (elevation 7,200m), but is absent in N. c. chilensis examined from Valparaiso Province. The modified segments of Neotyphloceras spp. are among the most complex in the Order, yet they have never been studied in detail sufficient to determine the validity of these sub-specific taxa. Sub-specific identification of females as N. c. chilensis was based on accompanying males. In the absence of accompanying males, the two females from the marsupial, T. elegans, could not be determined with certainty. Host specificity is of no value in differentiating Neotyphloceras ssp., as each is catholic in their host preference of cricetid and octodontid rodents and occasion- ally marsupials. Reports of Delostichus coxalis and T. corfidii have been restricted to octodontid rodents in the coastal montane environs of Valparaiso Province. Such limited distribution of these rather common species reflects a lack of collection effort, as their host species are much more widely distributed. Delostichus incisus has not been reported since its original description by Beau- cournu and Torres-Mura (1987) from two specimens of O. mimax (29 d, 49 9). Octomys mimax is restricted to the western Andean foothills and montane slopes of western Argentina. Although additional collecting of fleas from O. mimax in areas outside the type locality (Parque Nacional Ischigualasto) will undoubtedly extend its range in Argentina, it is likely not present in Chile. Beaucournu and Gallardo (1991, 1992) summarized 91 species of fleas in Chile, whereas one additional species (Ceratophyllus altus) was reported by Hastriter (2001). In summary, the description of two new species document 31 species in the genus Ectinorus and a total of 94 species reported in Chile. 178 Annals of Carnegie Museum VOL. 70 Acknowledgments Special thanks to Milton H. Gallardo and Fredy Mondaca, Institute de Ecologia y Evolution, Univ- ersidad Austral de Chile, Valdivia, Chile, who collected and generously provided the fleas for this study, which would otherwise not have been possible. Gratitude is also expressed to Theresa Howard and the Trustees of the Natural History Museum, London for the loan of flea specimens, to Robert E. Lewis, Ames, Iowa for the loan of specimens and permission to report previously unreported Argentine records of specimens from his collection, and to Michael F. Whiting and the staff of the Monte L. Bean Life Science Museum for their continued support in providing work space, equipment and materials necessary for conducting systematic studies. Literature Cited Beaucournu, J. C., and M. H. Gallardo. 1991. Siphonapteres du Chili; description de quatre especes nouvelles (Siphonaptera). Bulletin de la Society entomologique de France, 96:185--203. ■. 1992. Catalogue provisoire des puces du Chili (Insecta; Siphonaptera). Bulletin de la Society Frangaise de Parasitologie, 10:93-130. Beaucournu, J. C., and J. C. Torres-Mura. 1987. Un nouveau Delostichus (Siphonaptera, Rhopal- opsyllidae) d’ Argentine. Bulletin de la Soci6t6 Frangaise de Parasitologie, 5:257-260. Hastriter, M. W. 2001. Ceratophyllus altus Tipton and Mendez (Siphonaptera: Ceratophyllidae) in Chile, with notes on the distribution of the Genus Ceratophyllus Curtis, 1832 in the Southern Hemisphere. Proceedings of the Entomological Society of Washington. (In Press). Jordan, K. 1936. Some Siphonaptera from South America. Novitates Zoologicae, 34:305-310. — . 1942. On Parapsyllus and some closely related genera. Eos (Madrid), 18:7-29. Jordan, K., and N. C. Rothschild. 1923. On the genera Rhopalopsyllus and Parapsyllus. Ectopar- asites, 1:320-370. Johnson, P. T. 1957. A classification of the Siphonaptera of South America with descriptions of new species. Memoirs of the Entomological Society of Washington, 5:1-299. Lewis, R. E. 1976. A review of the South American flea subgenus Ectinorus Jordan 1942, with descriptions of two new species and a key (Siphonaptera: Rhopalopsyllidae). Journal of Parasi- tology, 62:1003-1009. Macchiavello, a. 1948. Siphonaptera de la costa sur-occidental de America (primera lista y distri- bucidn zoo-geogrdfica). Oficina Sanitaria Panamericana Publication 237, 1-49 (Washington, D. C.). Rothschild, M., and R. Traub. 1971. A revised glossary of terms used in the taxonomy and mor- phology of fleas. Trustees of The British Musuem (Natural History), London, United Kingdom. Rothschild, N. C. 1904. Further contributions to the knowledge of the Siphonaptera. Novitates Zoologicae, 1 1 : 602-65 3. — . 1909. On some American, Australian, and Palearctic Siphonaptera. Novitates Zoologicae, 16:61-68. — . 1914. New Siphonaptera from Peru. Novitates Zoologicae, 21:239-251. Smit, F. G. a. M. 1968. Siphonaptera taken from formalin-traps in Chile. Zoologischer Anzeiger, 180:220-228. — . 1987. An illustrated catalogue of the Rothschild collection of fleas (Siphonaptera) in the British Museum (Natural History). Vol. VII, Malacopsylloidea (Malacopsyllidae and Rhopalop- syllidae), Oxford University Press, The British Museum (Natural History), Oxford and London, United Kingdom. vS'f, ■•■. • t¥ • m »- ■ IJ. . t^x’*<>v; - HP*!* ••. VA&'v! W5Wi , ■■*::■ ?Vr*|It<(-*'’^' . -Ji-**? r*’ ' ,i V .' 'i| ■' ■'•^•' • J*^ '► ir^p^t .^IVi’irr ,,?f.. ■vr ; -ft '“ t u '..'< ■ #.'v' iiJ' -P'.-nlrf 4 r«*^ *.'.»■ .'.„ |y t '1 Ant.r <- ‘ ■>] ■ * i - « .1 k - t„rt.T t I ■. ■ » .-■:. ui'‘- .■'’V'S-'i ■> , »'i . •< l -f r'.M l- ., , :,;^- - iv >#•*■■(*) L, ♦ , •4<.* 'i * ■-'*■ 1- ■■., V, (A !*■. V>- J'ji'v ‘ ,^■ .^' *' ■* '*'' ' yfc" ' ’Cl -t. ^ ^ ’ 4 ^ ■ ,t"'' V-^w: " ■' u‘ -fov i-rr# ^ ‘V ■ C .,CII V* : 'V r .' ■ ■y-ji 'mk' ■ ■■ • ,4^ . r^-si V iiJL ■ * ' -i m % ■ i-tci'it -“^l i. w J/’’' 2**^'. ^ ^ . 5? in 'J!ii $«cm: •'■•' ''’-'Mr „ , '^^Mi T • = air. '> ?- ■*> .V ». , ., "" . .'iU V '« *’ /V'’ r-f - .1 ". v%' -TWj C»r i>QW I '-^kttVjrsfSiioi:. i -!V.- . , i V ii.Jtli. -.,t. # Wi» , .-... , i-?-. « ;i *^f *';:<• ‘ * “f ^ **■•’• • -' v^yl ,iifji {itK ■’ •■ r •. .•ViVtiiCC* [pv - 1 .• ■ %*«- Sr-'^f nyt.1 ■■ l', i >: 1. :;-. .:.: *fy^'C '' 2 *J%&> fe-.i , ltd ‘ r,M‘^kL i INSTRUCTIONS FOR AUTHORS ANNALS OF CARNEGIE MUSEUM consists of contributions to the earth sciences (including paleontol- ogy), life sciences, and anthropology, in 30 by 46 pica format (127 by 195 mm or 5 by 7% inches). Submit all manuscripts to the Office of Scientific Publications. Au- thors should give particular attention to scientific con- tent, format, and general style for the ANNALS. Manu- scripts that do not conform to the style of the AN- NALS will be returned to the author immediately. Every manuscript will be subjected to external peer re- view. Authors should submit the names, addresses, phone and fax numbers, and e-mail addresses of at least four qualified potential reviewers for each manuscript. Authors will be asked to subsidize, if funds are available, any or all costs of publication (approximately $ 100/page printed). A detailed set of instructions to authors may be requested from the Office of Scientific Publications. Manuscript Style. — Articles should include the fol- lowing items in this order: title page, abstract, text (with desired headings), acknowledgments, literature cited, ta- bles, figure captions, and copies of illustrations. All manuscripts must be typed double-spaced on standard 8%- by 1 1-inch white bond paper, with at least one-inch margins all around, and submitted in quadruplicate — an original for the editors, and three review copies. All pages should be numbered, including tables, literature cited, and the list of figure captions. Only correspon- dence quality or better dot matrix printouts will be ac- cepted; draft mode is unacceptable. A monospaced font, (e.g.. Courier) is required. A disk in WordPerfect x.x or ASCII is helpful. Title Page. The title should be brief, include the ani- mal or plant group involved, be placed on a separate page, and appear two to three inches below the top mar- gin. Include the author’s name(s) and the affiliations of non-CMNH authors. In the case of multiple authorship, indicate the address to which proofs should be sent. Abstracts. Abstracts should be short, but substantive, and included at the head of the first page of text. Text. Do not right justify text or break (hyphenate) a word at the end of a line. Footnotes and acknowledg- ments as footnotes are unacceptable. All text is double- spaced in a monospaced font (e.g.. Courier). Tables and Figure Legends. The list of figure legends and each table should be typed consecutively on indi- vidual pages separate from the text. Tables must be dou- ble-spaced throughout with no vertical lines. Measurements. Metric units should be used, if possi- ble. Literature Cited. All references to literature in text must appear in the Literature Cited section. The data (author, date, and page) in both citations must agree. Do not abbreviate the titles of periodicals or serials. The following style, double-spaced, should be used in Lit- erature Cited: 1) Two authors in a journal series: Soltis, D. E., and P. S. Soltis. 1992. The distri- bution of selfing rates in homosporous ferns. American Journal of Botany, 79:97—100. 2) Same authors repeated — use three-em dash: . 1923. The fauna of the Ardyn Obo For- mation. American Museum of Natural History Novitates, 98: 1-5. 3) Same authors plus a third author — repeat all authors: Knutson, L. V, R. E. Orth, and W. L. Murphy. 1986. Catalog of Sciomyzidae (Diptera) of America north of Mexico. Entomography, 4: 1-53. 4) Chapter in an edited volume: Rausch, R. L. 1963. A review of the distribution of Holarctic mammals. Pp. 29^3, in Pacific Basin Biogeography (J. L. Gressitt, ed.). Bishop Museum Press, Honolulu, Hawaii. 5) Unpublished dissertation: Smith, J. P. 1976. Review of Eocene Mammals. Unpublished Ph.D. Dissert., University of California, Berkeley, California. 6) Book: White, M. J. D. 1961. The Chromosomes. Meth- euen and Co., Ltd., London, United Kingdom. 7) Journal articles with usual volume and issue number: Anderson, W. I. 1969. Lower Mississippian con- odonts from northern Iowa. Journal of Pale- ontology, 43:916-928. Figures. Including all illustrative materials (line art, halftones, photographs), figures are to be numbered in Arabic numerals. Four sets of figures are required, one (original artwork) for reproduction, three for reviewers. Photocopies of photographs for reviewers are usually not acceptable but are adequate for line drawing review cop- ies. Figures may not be larger than 17 by 12 inches. Reducing figures is the responsibility of the author. All figures must be reducible to a maximum of 127 by 195 mm (30 by 46 picas) without loss of clarity. Line copy should be designed for reduction to % or Vi or actual size. Typewritten figure copy will not be accepted. Pho- tographic figures should be submitted at actual repro- duction size, if possible. Rectangular halftone figures should be abutted, with- out intervening spaces. The printer will insert narrow white spaces during the reproduction process. All figures must have minimally one-inch borders all around. Each figure should be given a protective cover and identified on the back side. Lettering and/or a magnification scale (linear metric scale) for rectangular halftone figures should be placed directly on the photo, not in a blank space between pho- tos. The scale or lettering for closely cropped photos can be placed in blank areas close to the figure. Proof. — The author should answer all queried proof marks and check the entire proof copy. Return corrected page proof with the edited manuscript promptly to the Office of Scientific Publications. . If an author chooses to make extensive alterations to a paper in proof stage, the author will bear the cost. Original manuscripts will not be returned unless request- ed. Figures will be returned to the author only if re- quested prior to publication. ISSN 0097-4463 AS 2(o P7X "Xn nals 0/ CARNEGIE MUSEUM THE CARNEGIE MUSEUM OF NATURAL HISTORY 4400 FORBES AVENUE • PITTSBURGH, PENNSYLVANIA 15213 VOLUME 70 28 NOVEMBER 2001 NUMBER 4 CONTENTS ARTICLES Leptotarsus (Tanypremna) in the Lesser Antilles: Description of a new species from Guadeloupe and biogeographical notes (Diptera: Tipulidae) Chen W. Young 239 Small mammals and Foraminifera from the Anatolian (Central Taurus) Early Miocene Engin Unay, E§ref Atabey, and Ger^ek Sara9 247 A new species of Gobiolagus (Lagomorpha, Mammalia) from the Middle Eocene of Shanxi Province, China Zhao-qun Zhang, Mary R. Dawson, and Xue-shi Huang 257 FROM THE ARCHIVES AND COLLECTIONS C. V Hartman’s letter of February 20, 1903 to W. J. Holland David R. Watters and Oscar Fonseca Zamora 263 Index to Volume 70 269 m 2 3 2002 Editors, ANNALS, BULLETIN and SPECIAL PUBLICATIONS: D. S. Berman, Ph.D. D. R. Watters, Ph.D. J. R. Wible, Ph.D. W. O. Goehring, Managing Editor Manuscripts, subscriptions, orders for individual numbers, and changes of address should be sent to: Office of Scientific Publications Carnegie Museum of Natural History 4400 Forbes Avenue Pittsburgh, PA 15213-4080 Phone (412) 622-3287 Fax (412) 622-8837 E-mail: scipubs@clpgh.org ANNALS OF CARNEGIE MUSEUM is published quarterly by The Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania 15213-4080, by the authority of the Board of Trustees of Carnegie Institute. © 2001 Carnegie Institute. @ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 4, Pp. 239-245 28 November 2001 LEPTOTARSUS (TANYPREMNA) IN THE LESSER ANTILLES: DESCRIPTION OF A NEW SPECIES FROM GUADELOUPE AND BIOGEOGRAPHICAL NOTES (DIPTERA: TIPULIDAE) Chen W. Young Associate Curator, Section of Invertebrate Zoology Abstract Leptotarsus (Tanypremna) guadeloupensis new species is described from wet secondary forests on the Lesser Antillean island of Guadeloupe. The external morphology of male and female genitalia of the new species and its closest relative, Leptotarsus (Tanypremna) hodgei Alexander from Dominica, are described and illustrated. The relationship between the two species and the possible origin of the subgenus Leptotarsus (Tanypremna) in the Lesser Antilles are briefly discussed. Key Words: Leptotarsus, Tanypremna, Tipulidae, biogeography, Guadeloupe, Antilles Introduction Crane flies in the genus Eeptotarsus Guerin-Meneville are largely tropical or warm temperate, with only a few species reaching their distributional limits in the Northern Hemisphere (Oosterbroek and Theowald, 1992; Alexander, 1965). About 300 described species of Leptotarsus exhibit a Gondwanian distribution (Gelhaus and Young, 1995), and are concentrated in two species^rich regions, one in the Neotropical Region (Alexander and Alexander, 1970), and the other in the Australian Region (Oosterbroek, 1989). Leptotarsus as currently classified is con- sidered to be doubtfully monophyletic (Young and Gelhaus, 1992) based on great morphological diversity. Twenty subgenera have been proposed within Leptotar- sus with Longurio being the most widespread, found in all major faunal regions except the Australian. Eleven subgenera with 95 species have been described in the Neotropical Region according to the Catalogue of the Craneflies of the World (Oosterbroek, personal communication), predominantly in South America. No subgeneric revisionary work has been done on the genus. Only a single species is known from the Caribbean area, Leptotarsus (Tanypremna) hodgei Alexander from Dominica in the Lesser Antilles. A second species of Leptotarsus (Tany- premna) on Guadeloupe is described below, and provides some explanation for the origin of Leptotarsus in the Lesser Antilles. Alexander treated Tanypremna Osten Sacken as genus when T. hodgei Alex- ander was first described in 1939. The status of Tanypremna remained as genus until 1945 when it and other related taxa were treated as subgenera of Leptotarsus. T. hodgei was later transferred to subgenus Longurio Loew in 1970 by Alexander without any comment. The only remark by Alexander was at the end of the description of the species L. (T.) clotho (Alexander, 1944), where he mentioned '"Longurio Loew and Tanypremna Osten Sacken are, in reality, all closely related and may well be found to pertain to a single major generic group.” However, T. hodgei was listed as Leptotarsus (Tanypremna) hodgei in the Catalogue of Neo- tropic Tipulidae (Alexander and Alexander, 1970). The new species is described Submitted 1 February 2001. 239 240 Annals of Carnegie Museum VOL. 70 here in the subgenus Tanypremna, and hodgei is maintained in Tanypremna for the following three reasons. First, after the description of hodgei, Alexander com- pared hodgei with four other species: carbonipes Alexander, A Alexander, invaripes Alexander, and kadeni Alexander, which are presently all still classified in the subgenus Tanypremna. Secondly, the Neotropic species of Leptotarsus seem to center around the subgenera Tanypremna and Longurio. Lastly, the sub- genera of Leptotarsus are in need of revision and it seems better to follow the classification for all Neotropical species of Leptotarsus as given in the Neotropical Catalogue (Alexander and Alexander, 1970), instead of Alexander, 1970, which deals with only one species. Systematic Entomology Order Diptera Linnaeus, 1758 Family Tipulidae Latreille, 1802 Subfamily Tipulinae Latreille, 1802 Leptotarsus (Tanypremna) guadeloupensis Young, new species (Fig. lA-B) Diagnosis. — Males of this species are extremely similar to that of L. (T) hodgei but differ by having a distinct dorsal spine on middle part of the inner dististyle. Female differs from L. (T.) hodgei by having the median extension of the posterior border of Tergum X extended medially reaching inner base at about one-half length of the paired cerci. In L. (T.) hodgei the extended Tergum X reaches inner base at about one-third length of the paired cerci. Description. — Body length: S, 13 mm; 9, 13 mm. Wing length: 6, 12 mm; 9,11 mm. Head: Occiput and rostrum reddish fulvous, slightly pruinose at vertex; rostrum short; nasus incon- spicuous; palpi yellow, with distal segment subequal in length to preceding ones combined. Eyes widely separated dorsally and ventrally. Antenna short, 13 segmented, subequal to length of palpi; scape cylindrical, twice as long as wide; pedicel globular; flagellomeres simple, with dorsal and ventral verticillar setae twice as long as corresponding flagellomeres; first flagellomere distinctly enlarged, narrowed at base, expanded distally; remainder of flagellomeres shorter than first and tapered apically. Thorax: Mesonotal praescutum yellowish brown, with a narrow median line, and two shorter, broad- er, fainted lateral stripes extending from prescutal pits to transverse suture; small, dark brown, trian- gular areas at prescutal pits, laterodorsal corners of scutum, and anterior of lateral prescutal stripes. Pleura evenly yellow, except for dark brown katatergite. Wings grayish subhyaline, with brown suf- fusion; subcostal cell and stigma dark brown; discal cell hexagonal; Rs one and half length of m-cu. Haltere well developed, with yellowish, cylindrical basal pedicel and dark, conspicuous blade-like capitulum. Legs long and slender in both sexes; coxae and trochanters yellowish brown; femora and tibiae brownish black with dark tips; tibial spur formula 1-1-1; tarsal claws each with one basal tooth and one middle tooth in both sexes. Abdomen: Abdomen elongate; tergum bicolored with apical half dull black and outer portion of basal half yellow with a narrow blackened ring at base; sternum yellowish brown, basal dark ring indistinct, basal yellow area extending into more than half of each segment; Tergum VIII and hypo- pygium uniformly black; setae scattered only over darker areas of terga. Hypopygium (Fig. lA): Posterior margin of Tergum IX emarginated. Basistyle long, slightly nar- rowed apically. Both dististyles elongate, fused at bases. Outer dististyle simple, spatulate, with setae of moderate length, fused with inner dististyle for two-thirds its length. Inner dististyle dilated at base, bent mesocaudally near middle forming elongate beak, beak with two small blackened spinoid setae on outer margin before apex, outer basal lobe long, curved inward from base into long, gradually recurved, black terminal spine; a second distinct spine, directed anteriorly, located on dorsal part in the middle of inner dististyle. Ovipositor (Fig. IB): Ratio of (length Tergum VIII + cerci) / (length Sternum VIII + hypovalves) = 1 .4 in cleared specimens. Tergum VIII rectangular, posterior border straight. Tergum IX not clearly separated by intersegmental membrane from Tergum X; length of Tergum IX one-half that of Tergum VIII; length of Tergum X three times that of Tergum IX. Posterior border of Tergum X extended 2001 Young — Leptotarsus Crane Flies in the Lesser Antilles 241 Fig. 1. — A, B. Adult of Leptotarsus (Tanypremna) guadeloupensis: A. male hypopygium, dorsal view; B. female ovipositor, lateral view. C, D. Adult of Leptotarsus (Tanypremna) hodgei: C. male hypo- pygium, dorsal view; D. female ovipositor, lateral view. E. distal end of hind leg, lateral view. Figs. A, B, C, and D same scale. medially, reaching about one-half length of inner base of paired cerci. Cerci of moderate length, dorsal surface flat, broad at base and narrowed to rounded apex. Outer surface of cerci glabrous, large setae scattered over basal half on ventral surface of cerci. Hypovalves basally with a dark brown, transverse band. Outer surface of valves glabrous, dark setae on inner surface of hypovalves near base. Furca of moderate length, extending to intersegmental region. Sternum IX (fused valvulae) with pigmented and sclerotized transverse band; medially with elongate, narrow blade, sclerotized and dark pigmented. Infra-anal lobe broad, rounded, with long setae scattered along apical half. Primary Type. — Holotype S (Zoological Museum of Amsterdam, The Neth- erlands). Verbatim label data: “GUADELOUPE: Basse Terre. Small streamlet in secondary forest near Maison du Volcan (La Soufriere), 950m above sea level, 242 Annals of Carnegie Museum VOL. 70 27 May 2000, L. Botosaneanu leg. Hypothermal water (27.5 C)” / “HOLOTYPE 6 Leptotarsus (Tanypremna) guadeloupensis C.W. Young.” Paratypes. — GUADELOUPE. Basse-Terre: Gourbeyre (“Moscou”), at light, 16 April 1992 (L. Botosaneanu), 1 $ (Zoological Museum of Amsterdam, Netherlands); Deuxieme Chute du Carbet, at light, 10 April 1992 (L. Botosaneanu), 1 S (teneral) (CMNH). Etymology. — The new species is named in reference to the type locality of this species, the island of Guadeloupe. Remarks. — Leptotarsus (T.) guadeloupensis appears most closely related to the only other Antillean species, L. (T.) hodgei from Dominica, based on the basal morphology of the male hypopygium, especially the large posterior spine on the outer basal lobe of the inner dististyle. Males of the new species can be easily distinguished morphologically from L. (T.) hodgei by having an additional distinct dorsal spine in the middle of the inner dististyle, and by having the outer dististyle fused more than half its length to the inner dististyle. Females of the new species have the posterior border of Tergum X extended medially, reaching inner base at about one-half length of the paired cerci. Leptotarsus (Tanypremna) hodgei (Alexander) 1939 (Fig. IC-E) Tanypremna (Tanypremna) hodgei Alexander, 1939:92-93. Leptotarsus (Longurio) hodgei (Alexander): Alexander, 1970:3. Leptotarsus (Tanypremna) hodgei (Alexander): Alexander and Alexander, 1970:14. Diagnosis. — This species can be distinguished from all other Tanypremna by the large posterior spine on outer basal lobe of the inner dististyle in male spec- imens. The female of this species is redescribed here to include the illustration of ovipositor. Description. — Body length: S, 14 mm; 9,18 mm. Wing length: 6, 12 mm; 9, 12 mm. Head: Occiput and rostrum reddish brown, slightly pruinose at vertex; rostrum short; nasus incon- spicuous; palpi yellowish brown, with distal segment subequal in length to preceding ones combined. Eyes widely separated dorsally and ventrally. Antenna short, 13 segmented, subequal to length of palpi; scape cylindrical, twice as long as wide; pedicel globular; flagellomeres simple, with dorsal and ventral verticillar setae twice as long as corresponding flagellomeres; first flagellomere distinctly en- larged, narrowed at base, expanded distally; remainder of flagellomeres shorter than first and tapered apically. Thorax: Mesonotal praescutum yellowish brown, with four nitidulous, dark brown stripes, the in- termediate pair separated by a long and slender darker vitta, the posterior interspaces faintly indicated; posterior sclerites of notum dark brown, sides of mediotergite pale brown. Pleura medium brown, without pattern. Wings grayish subhyaline, with brown suffusion; subcostal cell and stigma dark brown; discal cell hexagonal; vein Rs one and half length of crossvein m-cu. Haltere well developed, with yellowish, cylindrical basal pedicel and darker, conspicuous blade-like capitulum. Legs long and slender in both sexes; coxae and trochanters pale brown; femora and tibiae brownish black with darker tips; tibial spur formula 1-1-1; tarsi black; tarsal claws each with one basal tooth and one middle tooth in both sexes (Fig. IE). Abdomen: Abdomen elongate; tergum bicolored with apical half black and outer portion of basal half yellow with a narrow blackened ring at base; black including more than the outer half of segments IV-VII; sternum yellowish brown, basal dark ring indistinct, basal yellow area extending into more than half of each segment; Tergum VIII and hypopygium uniformly black; setae scattered only over darker areas of the terga. Hypopygium (Fig. 1C): Posterior margin of Tergum IX emarginated. Basistyle long, slightly nar- rowed apically. Both dististyles elongate, fused at bases. Outer dististyle, simple, spatulate, with setae of moderate length, fused with inner dististyle about one-half of its length. Inner dististyle dilated at base, bent mesocaudally near middle forming elongate beak, beak with one or two small blackened spinoid setae on outer margin before apex (only one spinoid setae was indicated on Fig. 1C); outer basal lobe long, curved inward from base into long, recurved, black terminal spine. 2001 Young — Leptotarsus Crane Flies in the Lesser Antilles 243 Ovipositor (Fig. ID): Ratio of (length Tergum VIII + cerci) / (length Sternum VIII + hypovalves) = 1 .0 in cleared specimens. Tergum VIII rectangular, posterior border straight. Tergum IX not clearly separated by intersegmental membrane from Tergum X; length of Tergum IX two-thirds that of Tergum VIII; length of Tergum X twice that of Tergum IX. Posterior border of Tergum X extended medially, reaching about one-third length of inner base of paired cerci. Cerci of moderate length, dorsal surface flat, broad at base and narrowed to rounded apex. Outer surface of cerci glabrous, large setae scattered over basal half on inner surface. Hypovalves basally with dark brown, transverse band. Outer surface of valves glabrous, dark setae on inner surface of hypovalves near base. Furca of moderate length, extending to intersegmental region. Sternum IX (fused valvulae) with pigmented and sclerotized trans- verse band. Medially with elongate, narrow blade, sclerotized and darkly pigmented. Infra-anal lobe broad, rounded, with long setae scattered along apical half. Primary Type. — Holotype 6 [on paper point pinned on same pin with paratype female] (NMNH). Verbatim label data: “Dominica Morne Trois Pi tons 4500' VIII-15-'38 W. FI. Hodge” [handwritten in black ink] / “HOLOTYPE S Tany- premna hodgei C. P. Alexander” [red paper label]. Other Material Examined. — DOMINICA. St. Andrew Parish: d’Leau Gommier, 17 March 1956 (J. F. G. Clarke), 1 3 (NMNH); d’Leau Gommier, 16 March 1965 (W. W. Wirth), 1 6 (NMNH). St. George Parish: Fresh Water Lake, 2500 ft, 5 April 1966 (R. J. Gagne), 1 <$ (NMNH). St. Paul Parish: Morne Trois Pitons, 4,500 ft, 15 August 1938 (Hodge), 2 cJ, 2 $ [paratypes of Tanypremna (Tanypremna) hodgei Alexander] (NMNH); Pont Casse, 23 November 1964 (P J. Spangler), 1 6 (NMNH); 0.5 km NE Pont Casse, NW slope of Trois Pitons, 15-22N, 61-21W, 560m, 12 June 1991 (J. E. Rawlins, S. A. Thompson), 2 cJ, 1 9 (CMNH); 0.5 mi E Pont Casse, 11 April 1966 (R. J. Gagne), 1 3 (NMNH); 3 mi E Pont Casse, 13-16 October 1966 (A. B. Gurney), 1 6 (NMNH). All specimens collected between 1964 and 1966 were from the Bredin-Archbold-Smithsonian Biological Survey of Dominica. Remarks. — The original description of L. (T.) hodgei was based on specimens collected from Morne Trois Pitons, Dominica (Alexander, 1939). Alexander (1970) provided supplementary remarks on variation when additional specimens became available, but he regarded all specimens as belonging to a single species. Examination of material collected in 1991 demonstrated further variation in the overall shape of the male inner dististyle and the number of apical spinoid setae ranging from 1 to 2 on the dististyle beak. However, I concur with Alexander that these are differences between individuals within a single species on Dominica. Discussion The crane fly fauna of the West Indies and its relationship to the continental fauna of South America and Central America is poorly understood. Geographical distributions of most tipulid species are inadequately documented. Few surveys have been conducted and only a limited number of specimens are available for examination. Previous studies have been isolated taxonomic papers (Alexander, 1936, 1931a, 1931b, 1939) or limited to the Greater Antilles (Alexander, 1932, 1964). Only two islands in the Lesser Antilles have been studied in some detail. Saint Vincent (Williston, 1896) and Dominica (Alexander, 1970). The biogeography of the Caribbean Tipulidae is beyond the scope of the present study without a complete phylogenetic analysis. However, a few tentative hy- potheses can be presented based on the newly discovered L. (T.) guadeloupensis, and from unpublished results from intensive field research on crane flies currently underway elsewhere in the Caribbean, especially Hispaniola and Puerto Rico. Two possible dispersal routes have been suggested for the arrival of insects to the oceanic islands of the Lesser Antilles (Liebherr 1989; Smith, Miller and Miller, 1994). One is northern and involves the Greater Antilles, particularly Puerto Rico and the Virgin Islands, and the other is southern from Trinidad and continental 244 Annals of Carnegie Museum VOL. 70 South America. Based on the currently known distribution of described species, the subgenus Tanypremna has a southern radiation with 31 species restricted to the Neotropical Region, most in South America. Of these, only two species are Caribbean, both in the Lesser Antilles on Dominica and Guadeloupe. No species of the subgenus is known from the Greater Antilles. Morphologically, L. (T. ) guadeloupensis and L. (T.) hodgei are more similar to each other than to any other mainland species in the subgenus. Both are characterized by a dilated inner dististyle with its outer basal lobe forming a long, curved, black spine, a feature not found in any other species of Tanypremna. This character is presumably apomorphic and supports the monophyly of these two species. It further suggests that speciation of Lesser Antillean Tanypremna occurred after the immediate com- mon ancestor diverged in isolation from a South American radiation in that genus. The present study confirms that Tanypremna species in the Lesser Antilles may be restricted to individual islands. L. (T.) guadeloupensis, with its diagnostic ad- ditional spines on the male inner dististyle, might be considered more derived than L. (T.) hodgei, and if so supports a scenario in which an ancestral population established from South America on Dominica diverged over time to form L. (T.) hodgei, and somewhere during that divergence a population dispersed to Gua- deloupe and further evolved to become L. (T.) guadeloupensis. But the loss of these additional spines in L. (T.) hodgei from Dominica has the same parsimony from a strict phylogenetic point of view and thus presents another equally likely scenario. A third scenario would be that the ancestral population gave rise to both species simultaneously. Testing these hypotheses requires the study of the tipulid fauna on adjacent islands in addition to further revisionary and phylogenetic work on crane flies in general. Tipulidae are not rigorously collected by non-specialists due to their fragility and this is especially true in the Lesser Antilles. It is likely that closely related species will be collected on neighboring islands such as Nevis and Antigua in the north, and Martinique and St. Lucia to the south. Distributional patterns of several other lineages of Tipulinae in the Caribbean region are dis- tinctly different from that of the subgenus Tanypremna, and they constitute chal- lenges for further research. For^example, the genus Nephrotoma, with a northern radiation, is found throughout the Antilles. The genera Brachypremna and Meg- istocera, both predominantly with southern radiations, are found in the Greater Antilles but not the Lesser Antilles. Of particular interest is the subgenus Doli- chopeza (Megistomastix), because of its high diversity and endemicity to the West Indies, primarily the Greater Antilles, and its likely sister group relationship with the Afrotropical subgenus D. (Trichodolichopeza). Much additional field collect- ing on all Caribbean islands, and further revisionary work on Caribbean crane flies, will be required to interpret these diverse distribution patterns in the West Indies. Acknowledgments Very special thanks are due Dr. Lazare Botosaneanu of the Zoological Museum of Amsterdam, The Netherlands. This study would be impossible without his specimens from the island of Guadeloupe. The 2000 expedition of Dr. Botosaneanu to Guadeloupe has been financed by the TREUB MAAT- SCHAPPIJ, Amsterdam, and by the PARC NATIONAL DE LA GUADELOUPE. Thanks are also due Dr. Pjotr Oosterbroek for suggesting this project and comments on an earlier draft. Additional specimens from Dominica were borrowed from the United States National Museum of Natural History, Washington, D.C., through the assistance of Nancy Adams and Holly Williams. I would like to thank John E. Rawlins, Carnegie Museum of Natural History, for suggestions on the manuscript; also to three anonymous reviewers for their comments. 2001 Young — Leptotarsus Crane Flies in the Lesser Antilles 245 Literature Cited Alexander, C. R 1932. The crane-flies of Puerto Rico. Journal of the Department of Agriculture of Puerto Rico, 16:349-387. . 1936. New or little-known species of West Indian Tipulidae (Diptera). I. Journal of Agri- culture of the University of Puerto Rico, 20:877-882. . 1937fl. New or little-known species of West Indian Tipulidae (Diptera). II. Journal of Ag- riculture of the University of Puerto Rico, 21:179-188. . 1937.Z? New or little-known species of West Indian Tipulidae (Diptera). III. Journal of Ag- riculture of the University of Puerto Rico, 21:523-534. . 1939. New or little-known species of West Indian Tipulidae (Diptera). IV. Journal of Agri- culture of the University of Puerto Rico, 23:91-130. . 1944. Records and descriptions of Neotropical Crane-flies (Tipulidae, Diptera), XVIII. Jour- nal of the New York Entomological Society, 52:369-383. . 1964. The crane-flies of Jamaica (Diptera, Tipulidae). Bulletin of the Institute of Jamaica, Sciences Series 14:1-86. . 1965. Family Tipulidae. Pp. 16-19, in A Catalog of the Diptera of America North of Mexico (Stone et ak, ed.). Agriculture Handbook No. 276. U. S. Department of Agriculture. . 1970. Bredin-Archbold-Smithsonian Biological Survey of Dominica. The crane flies (Dip- tera: Tipulidae). Smithsonian Contributions to Zoology, No. 45:1-59. Alexander, C. P., and M. M. Alexander. 1970. Family Tipulidae. Fascicle 4:1-259, in Catalogue of the Diptera of the Americas South of the United States (N. Papavero, ed.). Museu de Zoologia, Universidade Sao Paulo. Gelhaus, J. K., and C. W. Young. 1995. Pupae of the crane fly genus Leptotarsus (Diptera: Tipulidae) in the New World, with discussion of the monophyly of the genus. Annals of Carnegie Museum, 64:135-145. Liebherr, J. K. 1989. General patterns in West Indian insects, and graphical biogeographic analysis of some circum-Caribbean Platynus beetles (Carabidae). Systematic Zoology, 37:385-407. Oosterbroek, P. 1989. Tipulidae. Pp. 53-116, in Catalog of the Diptera of the Australasian and Oceanian Regions (N. L. Evenhuis, ed.). Bishop Museum Press, Honolulu, and E. J. Brill, Leiden. Oosterbroek, R, and B. Theowald. 1992. Family Tipulidae. Volume 1:56-178, in Catalogue of Palaearctic Diptera (A. Soos, L. Papp, and P. Oosterbroek, eds.). Hungarian Natural History Mu- seum, Budapest. Smith, D. S., L. D. Miller, and J. Y. Miller. 1994. The butterflies of the West Indies and South Florida. Oxford University Press, Oxford, United Kingdom. WiLLiSTON, S. W. 1896. On the Diptera of St. Vincent (West Indies). Transactions of the Entomological Society of London, 1896:253-446. Young, C. W., and J. K. Gelhaus. 1992. Leptotarsus (Longurio) byersi, a new flightless crane fly from Ecuador (Diptera: Tipulidae). Acta Zoologica Cracoviensia, 35(1):97-105. . JM, ’ , 'a '. ^ -.‘■■^^ i-.^ <\/ •-- - -- .. * " J f ' '! /'• ‘'Vr.i >fr,, / V .,: -T : .V .■• ..,,, ,.:V ..■:.:...i,A tP 4 v\-.i‘U/>r ■*“ ■-'.'•if ' ;,y '’q-s it-Tr/rf ^1^',, '••. :/ A* >’wj3f .%C; ^ f £'^- ■^'V^v i ■'’ '■’jJ^4^ i ' . * ' ■'mil'- - ' ' ' . i' i • ' ' ■ ■ ■ •;* ■;'■'■;>? '■’ >•;:-. 'jj;. ?? '’'VV|k|^ - , , ^ .0 • 1 «rfri -I .■t'j"V ■ . i >*., !► '.■• ^ -“i ■ w’* ‘ <1 r> j.^-.»K :* •' '■ c^ \yi^ sn,:- >; .S^9a4i■,rY•^u^-/><^^^«^. ; {r-> ■ ■ : .^.®t ^ ' *-■' ■ .. ciS' , •1’' rt Ja . f ' ”■ y ^f' ' :' " ''3. 1 ■!“T 3 ■ ■■— -a* li- ,» f ,-|«^ '-■*^; ... ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 4, Pp. 247-256 28 November 2001 SMALL MAMMALS AND FORAMINIFERA FROM THE ANATOLIAN (CENTRAL TAURUS) EARLY MIOCENE Engin Unay^ E§ref Atabey^ GeR^EK SARAg^ Abstract The age of the terrestrial elastics of the Deringay Formation that outcrop widely in the Mut region is controversial. In an attempt to resolve their age, small mammals have been collected at the locality Yapmti from the lacustrine mudstones in this formation. This assemblage contains Enginia cf. E. gertcheki, Cricetodon cf. C. kasapligili, Cricetodon aff. C. aliveriensis, Megacricetodon coUongensis, Democricetodon sp., Karydomys n.sp., Eumyarion sp., Tamias sp. Gliridae gen. et sp. indet, Sayimys sp., Albertona n.sp., Galerix uenayae, Talpidae gen, et sp. indet, Oligosorex sp., Soricidae gen. et sp. indet. This association suggests an Early Miocene Age (MN3 to early MN4) which roughly correlates with the middle Burdigalian (±18 Ma). The Yapmti assemblage is of special interest because the continental deposits of the Deringay Formation that yielded the rodents are disconformably overlain by the marine Mut and Kdselerli formations assigned to the late Burdigalian on the basis of planktonic Foraminifera, which is a first order correlation between continental and marine deposits. Key Words: Rodents, Lagomorphs, Insectivores, Early Miocene (MN3/4), Mut Basin (Turkey) Introduction The small mammals from Yapmti were collected in the context of the project ‘Sequence stratigraphy of Mut Miocene Basin’ that was carried out by Maden Tetkik ve Arama genel Mudiirliigu (MTA) in 1999. The study area was in the western part of Mut Basin, Central Taurus, Turkey (Fig. 1). The continental clastic sediments of the Deringay Formation are exposed in the Goksu River valley and around the Firing Suyu River (Fig. 2). The age of these elastics is controversial: Gedik et al. (1979), Uguz (1989), and Tanar and Gokgen (1990) attributed these sediments to the Burdigalian on the basis of benthonic and planktonic Forami- nifera; Bilgin et al. (1994) attributed them to the Late Oligocene-Lower Miocene on the basis of Ostracoda; and Atabey et al. (2000<2) attributed them to the middle Burdigalian on the basis of planktonic Foraminifera. This uncertainty over the age of these elastics had its effect on the age assignment of the overlying marine rock units. This study attempts to date the terrestrial elastics of the Deringay Formation using small mammals collected by wet screening three tons of sedi- ment, The measurements of the teeth are given in millimeters, with the first mea- surement being anteroposterior and the second being transverse. ' Cumhuriyet Universitesi, Fen-Edebiyat Fakiiltesi, Antropoloji Boliimii, 58140 Sivas, Turkey. 2 Maden Tetkik ve Arama (MTA) Genel Mudurlugii, Jeoloji Etiitleri Dairesi, 06520 Balgat-Ankara, Turkey. Submitted 5 July 2001. 247 248 Annals of Carnegie Museum VOL. 70 Fig. 1. — Location map of the Mut region with exploded view of the study area. 2001 Unay et al. — Anatolian Early Miocene Mammals and Foraminifera 249 EXPLANATIONS I NORTH 0 ‘ 1 2 Km Koselerli fm (Marl ) 1 upper Burdigalian- Mut fm. (Limestone) J an ^ UNCONFORMITY Derincay fm. (conglomerate, sanstone, mudstone) Middle Burdigalian Fakirce fm. :(c1aystone) ? Aquiatian-Lower Burdigalian % UNCONFORMITY upper cretaceous (Limestone) Fossil locality Fig. 2. — Geological map of the study area. Lithostratigraphy The clastic rocks of the Deringay Formation (Gedik et ak, 1979) are widely exposed in the study area. This unit (Fig. 3) is composed of meandering river deposits containing gray-coloured, coarse-grained channel deposits and red mud- stones. The red mudstones are interpreted as flood plain deposits. Gray-coloured, gastropod-bearing mudstones, deposited in residual lakes, occur locally below and above the mudstones. The sandstones are laterally continuous but vary in thick- ness. The sandstones/conglomerates and mudstones intercalate. These elastics are disconformably underlain by the unfossiliferous lacustrine claystones, shales, 250 Annals of Carnegie Museum VOL. 70 E, CO H- Q C/5 oo LU LITHOLOGY EXPLANATIONS UJ :s^ "ZC SERIE STAGE cc o LL sc U 5 o z LU ULi 2 (J _ Kyl/^ri llmocf^^no a. UJ o o Q z o ^ m 0 S i m iViLIH ,L’)L4y^y JTTORAL/DE MARINE 5 UJ CO o <8 Q Q ^ 1 ^ 1 ^ i /Ci. 1 - , , , Foraminifers Limestone S A -r-^- |-‘ -A -1 -* J L__!_J 1 L_jL — lSt Sandy marl UNCONFORMITY 3 ■ ■ ■ ■; . Gray sandstone Red mudstone Z Channel deposit sandstone conglomerate 1 < o Flooding sandstone UL lO . X . . . . A. \ . 0 < cO -ZX'. A - LU 0 $ Sandy-silty gray mudstone Z cr '-.tf ,,' ^ %>’-'■ ■ ' • <-• w Y‘* .i. ' , ■ : ’“ - ' ., ‘ ■. - ■ ; ■^•' i.-’r ’;/'■*:! ♦ uA.i ■ 'Vvi-; < ’ •* ,-^ ‘ t '■ . h y 7 '4i • *■>'. '."'’ll'-' i- iw'f .-T- >.%' irr,.x. i-T,> n^r \ ■ ■ '. . :M-- J^} 1 iv ' ' , l (ii ‘ ' fi^., r. ' ; '* . ^ ; . * .1 1 ' ;r . *'* * \.-.mr' -••> J , /n»* . nil k#. "i > * V . V ^ ,.j- 4- jt : : -. : * < • ( ,r I .55. ., ■ S'-'Hir^i^sn I . rrM'f 'itJt;. ■-: • ■-Vy- • v , .'.-.v.y- * . .. "' '\ .fc . -L . 1* . '■ * .. ■ - : : ,- ■ * , ^ Y' ■ 4 ■» 7 . * T,v ■ '' ‘YP'* r ’v '- ' £-** • i -^r*" H.y..;.r;iViSt» i>6jl(^,, -s ' - ■ ■ „ . ' .i^i- ■ V ' '■ ■ •.-•-■'^^'r^^ '■ ' . ■■■■ . ■■ ■ ' •„ ^•' ^ ' ■ '%«'. . 1 - ~ . 7 -ic. ' , ^ ^ if»''4r^ %■ i 'itf ^AH||«i.i ^ *i- ^ ?► < ' • ' • ■ * • ? 'o frilly:. i-* Vv/.S'^ s.'“ s r; , -', : «•, w v'H^.isi^- . -‘fe v.-> ?•■.,- ^7 rTr ANNALS OF CARNEGIE MUSEUM VoL. 70, Number 4, Pp. 263-268 28 November 2001 FROM THE ARCHIVES AND COLLECTIONS C. V. HARTMAN’S LETTER OF FEBRUARY 20, 1903 TO W. J. HOLLAND David R. Watters' Oscar Fonseca Zamora^ The Context of the Letter Director W. J. Holland hired Carl Vilhelm Hartman as Carnegie Museum’s first Curator of the Section of Ethnology and Archaeology on February 28, 1903, a position Hartman held until May 1, 1908. Their negotiations progressed rapidly and concluded only one month after Hartman inaugurated contact from New York City. In his letter of January 28, Hartman summarized his previous field experi- ence and inquired whether Carnegie Museum might be “ . . , willing to engage my services for arch. & ethnological explorations in Spanish America.” Holland responded on February 10, informing Hartman “ . . . that possibly you may be the man to assume the duties of such a position.” However, Holland prudently wanted first “ ... to ascertain from you what practical experience you have had in Museum administration ...” and therefore required of Hartman a “ . . . sketch giving an account of your early training and of your career hitherto . . . .” Hart- man’s equally circumspect reply of February 15 acknowledged Holland’s “hon- orable proposition” and requested a day or two to consider the question. On February 20, Hartman fulfilled Holland’s request in a six-page, hand-written letter, most of which dealt with his museum background. The Content of the Letter Hartman’s carefully crafted letter is reproduced below in its entirety except as noted. 127 West, 83‘-d St. New York Febr. 20* 1903. Doctor W. J. Holland. Dear Sir. In reply to your valued letter of the 10* of Febr. I was only able the other day to write you a few lines. Allow me now to answer your ques- tions, to give you a short sketch of my early training and some infor- mation about my knowledge of museum work. ' Curator, Section of Anthropology. ^ Research Associate, Section of Anthropology (and Professor, University of Costa Rica). 263 264 Annals of Carnegie Museum VOL. 70 I was born in Orebro in Sweden [in] 1862. My father Doctor Carl Hartman was a professor of natural history, author of the “Flora of Sweden and Norway.” Having finished my seven years course at the college I first spent two years as a bookkeeper and student at an agri- cultural school. I then entered into the service of the Academy of Ag- riculture and devoted myself to various branches of applied botany. Hav- ing obtained a stipend of the Academy of Science for the pursuit of studies at some of the foremost botanic gardens of Western Europe, I spent six years in Copenhagen, London, Paris, Erfurt, Berlin, Amster- dam, etc. When Dr. Carl Lumholtz (who is a Norwegian) organized his expe- dition for the exploration of Sierra Madre he engaged my services as a botanist. Soon however I became greatly interested in the archaeological and ethnological features and as Dr. Lumholtz needed a special assistant, I turned my attentions wholly to this line of investigation. After the conclusion of Dr. Lumholtz’ first expedition I went with him to the Columbian Exhibition at Chicago, where I remained six months in the Anthropological Department, arranging exhibits and assisting in the pur- chase and packing of ethnological collections. Returning to Sweden I received a favorable offer to carry out explo- rations of my own in Central-America under the auspices of the Anthrop. Geogr. Society. This position I accepted and carried on explorations, which extended over a period of three years. On my way back I spent six months in the principal museums of the United States studying the archaeological collections. In Stockholm I undertook new training in the various duties of a mu- seum worker under the immediate supervision of Professor Hjalmar Stolpe. I classified, arranged and labelled anew my collections and wrote de- scriptive catalogues. At the same time I had the opportunity day after day of following in detail the work Dr. Stolpe was doing in the different departments, at that time especially valuable and instructive, on account of the complete reorganization of the institution then going on, according to thorough scientific and systematic plan. Although the experience thus gained during these three years of daily intercourse with Professor Stolpe does not seem to cover a very long time I consider the same really valuable, because my instructor is un- doubtedly one of the foremost museum men of Europe, having worked some 30 years inside the walls of the National Museum in Stockholm in association with Prof. Montelius and Hildebrand and having studied nearly all the museums of Europe, North & South America and Asia. Although the great life work of Stolpe has been principally in the domain of Scandinavian archaeological research, he is known in Europe as an authority on general ethnology also. During a tour around the world with one of the Swedish Princes he made extensive collections, especially in Japan, China, India and South- America. His principal study however has been that of the peoples of the Pacific Islands, particularly their orna- mental art. 2001 Watters and Fonseca — From the Archives and Collections 265 [Not reproduced here are three paragraphs in which Hartman includes quota- tions about Stolpe’s eminent status in the museum field. They include comments about Stolpe’s work by Prof. Brinton (from Science), Prof. A. B. Meyer (in Ger- man from Globus), and Prof. Alfred Haddon (in Evolution of Art). They were included to bolster Hartman’s arguments about his training under Stolpe.] These quotations I have given only for the reason, that you may un- derstand, that although as said my training inside the museum walls has been somewhat short, I have however had the opportunity of getting the best instruction possible. Even during the three months, which I passed last fall in the museums of the United States in the company of Dr. Stolpe, I learned a good deal. I am also assured of the most hearty cooperation by my former superior in anything I undertake. He will always be glad to give me advice and information. As I have only today been offered definitely still another position from the [American ?] Museum of Natural History to go into the field and I am desireous [sic] of coming to a decision, I must beg of you a speedy reply, if I am to consider a position in the Carnegie Museum, with a definite offer on your part as to time, salary, opportunities etc. Hoping that you will favor me in this I remain very sincerely yours C. V. Hartman [Letter, February 20, 1903, C. V. Hartman to W. J. Holland, Holland Archives, Box 8, file 53] In crafting this letter, Hartman emphasized his museum experience under Hjal- mar Stolpe in Sweden at the expense of a more detailed presentation of his over- seas fieldwork in Mexico with Lumholtz and later in Central America. Hartman did so because Holland had requested information specifically about his “practical experience” in museums. Moreover, in his initial letter of January 28, Hartman already had provided Holland with more detailed information about his fieldwork and his proposition for additional research in Spanish America. The Swedish expedition he had led to Central America lasted three years (1896-1899) and included seventeen months of archaeological research in Costa Rica followed by an almost equivalent span of ethnographic research in El Salvador and Guatemala. This Central American expedition nominally was sponsored by the Swedish So- ciety for Anthropology and Geography. The validity of Hartman’s comment about having been offered a position by the Museum of Natural History is questionable. The Significance of the Letter The letter of February 20 confirmed Holland’s belief in Hartman’s suitability for the position of Curator. Hartman’s letter and his “excellent testimonials” jus- tified, in Holland’s view, his decision to bring the matter to the Museum Com- mittee during its monthly meeting on February 28, 1903. The Museum Committee chaired by C. C. Mellor approved Holland’s recommendation, and that same day Holland and Hartman (who had come to Pittsburgh) completed their negotiations and signed the hiring document. Hartman reported for duty on March 17 and 266 Annals of Carnegie Museum VOL. 70 Fig. 1 . — Carnegie Museum of Natural History possesses two photographs (CM Neg. No. 4000 and 4001) of C. V. Hartman in the Pittsburgh area, both posed the same way and in the same woodland setting. Hartman’s face is somewhat blurred in both images because he slightly moved his head each time the negative was exposed. His face is clearer in this image (Neg. No. 4000), reproduced here for the first time. Hartman’s attire and pose suggest a formal photograph, perhaps even an official staff photograph (implied as well by its presence in the Archives), but the wooded setting seems incongru- ous. The key to the locality may reside with the photographer. Written on the back of the photograph is the following text: “C. V. Hartman, Curator of Archaeology, Pa., Alle[gheny]. Co[unty]., Powers Run Ravine, May 1907, O. E. Jennings photo.” Otto E. Jennings was Assistant Curator of Botany. The 2001 Watters and Fonseca — From the Archives and Collections 267 within two weeks was headed to Costa Rica to begin the seven-month archaeo- logical expedition for Carnegie Museum. Holland and Hartman met first in October 1902, a mere three months before the January 28 letter. Both had attended the 13^'" Session of the International Congress of Americanists (ICA) held at the American Museum of Natural History in New York City. The 13‘^ ICA proceedings volume (1905:lxiv) discloses that Hartman presented a paper, “Archaeological Researches in Costa Rica,” in which he summarized the results of the archaeological portion (May 1 896-September 1897) of the Swedish Central American expedition. Holland’s address to the ICA (I905:xxxi-xxxii) expounded upon his vision of a new Carnegie Museum. Immediately following the Congress in October, Hartman visited Pittsburgh as a member of a delegation of foreign dignitaries traveling to museums in various cities. Holland hosted the stopover in Pittsburgh and took advantage of that op- portunity to again advocate his vision of a new Carnegie Museum. Hartman (Fig. 1) referred to that visit in his January 28 letter to Holland, noting he had “ . . . the opportunity to admire your institution and heard you tell us about your great plans for the future.” Thus, the Hartman and Holland letters of January and February 1903 were communications between individuals acquainted with one another, each of whom already had some measure of the other’s background, interest, and intent. Neither party entered into the negotiations blindly. However, what was probably unknown to either party, at least initially, was that each was not the first choice of the other. Hartman documents at Carnegie Museum of Natural History and others held in the University of Pennsylvania Museum Archives disclose that Hartman had tried unsuccessfully to secure employment in museums in New York City, Washington, DC, and Philadelphia before approaching Carnegie Museum. Holland, meanwhile, had tried unsuccessfully to bring Stewart Culin from Philadelphia to become Curator in Pittsburgh. The hiring, therefore, was a timely and expedient event for both parties because, by February 1903, Holland had need of a Curator of Eth- nology and Archaeology and Hartman needed a job in order to resume his ar- chaeological research in Costa Rica. Acknowledgments Throughout Hartman’s tenure, the institution was known as Carnegie Museum, only later being changed to Carnegie Museum of Natural History (CMNH). Documents used in this study are found in the CMNH Hartman Archives and Holland Archives. The Adrienne and Milton Porter Charitable Foundation graciously supported this research project. The authors express their gratitude to CMNH Research Associate Hazel Johnson for tracking information in local newspapers about the delegation visiting Pittsburgh; CMNH Librarian Bernadette Callery and Anthropology Collection Manager Deb- orah Harding for advice on accessing archives under their control; and Reference Archivist Alessandro Pezzati for granting permission to examine the Hartman documents in the University of Pennsylvania Museum Archives. The Carnegie Museum of Natural History maintains a C. V. Hartman Web site at http://www.carnegiemuseums.org/cmnh/hartman. pertinent Annual Report (Carnegie Museum, 1908:14) discloses that Jennings was then making local collecting trips with different members of the staff. Hartman, because of his continuing interest in botany, may have accompanied Jennings on a collecting trip to Powers Run Ravine (ca. 12 km from Carnegie Museum) and been photographed at that locality by his friend. 268 Annals of Carnegie Museum VOL. 70 Literature Cited Carnegie Museum. 1908. Eleventh Annual Report of the Director for the Year Ending March 31, 1908. Carnegie Museum, Pittsburgh, Pennsylvania. Holland, W. J. 1905. [Address about Carnegie Museum to the 13“" Congress]. Proceedings of the International Congress of Americanists, 13‘^ Session, New York, 1902:xxxi-xxxii. Eschenbach Printing Co., Easton, Pennsylvania. International Congress of Americanists. 1905. Proceedings of the International Congress of Amer- icanists, 13“^ Session, New York, 1902. Eschenbach Printing Co., Easton, Pennsylvania. INDEX TO VOLUME 70 CONTENTS ARTICLES Late Eocene-Oligocene nonmarine mollusks of the Northern Kishenehn Basin, Montana and British Columbia Harold G. Pierce and Kurt N. Constenius The skull of Mesenosaurus romeri, a small varanopseid (Synapsida: Eupelycosauria) from the Upper Permian of the Mezen River Basin, northern Russia Robert R. Reisz and David S Berman A diadectid (Tetrapoda: Diadectomorpha) from the Lower Permian fissure fills of the Dolese quarry, near Richards Spur, Oklahoma Robert R. Reisz and Tammy E. Sutherland Rhizomyidae from the Lower Manchar Formation (Miocene, Pakistan) Wilma Wessels and Hans de Bruijn Fleas (Siphonaptera; Ctenophthalmidae and Rhopalopsyllidae) from Argentina and Chile with two new species from the Rock Rat, Aconaemys fuscus, in Chile Michael W. Hastriter A revision of the moth genus Leucania Ochsenheimer in the Antilles (Insecta: Lepidoptera: Noctuidae) Morton S. Adams Middle Eocene Ischyromyidae (Mammalia: Rodentia) from the Shanghuang fissures, south- eastern China Mary R. Dawson and Banyue Wang [Review of] Trogons and Quetzals of the World (Paul A. Johnsgard) Alejandro Espinosa de los Monteros Remarks on the Annals of Carnegie Museum's centennial and the introduction of a new feature W. Orr Goehring Editorial from Annals of Carnegie Museum, Volume 1, Number 1 (1901) William J. Holland An excavation in Guanacaste province, Costa Rica David R. Watters and Oscar Fonseca Zamora Leptotarsus (Tanypremna) in the Lesser Antilles: Description of a new species from Guadeloupe and biogeographical notes (Diptera: Tipulidae) Chen W. Young Small mammals and Foraminifera from the Anatolian (Central Taurus) Early Miocene . Engin Unay, E§ref Atabey, and Gergek Sarag A new species of Gobiolagus (Lagomorpha, Mammalia) from the Middle Eocene of Shanxi Province, China . Zhao-qun Zhang, Mary R. Dawson, and Xue-shi Huang C. V. Hartman’s letter of February 20, 1903 to W. J. Holland David R. Watters and Oscar Fonseca Zamora 1 113 133 143 169 179 221 231 233 235 237 239 247 257 263 269 270 Annals of Carnegie Museum VOL. 70 NEW TAXA NEW SUBFAMILIES tMycterosaurinae, new subfamily 114 NEW GENERA AND SPECIES '\Anatoparamys, new genus 222 '\Anatoparamys crepaturus, new species 222 tBiomphalaria spira, new species . 72 '\Cincinnatial bowmanana, new species 60 tCoelostemma dawsonae, new species 43 Ectinonis (Ectinorus) gallardoi, new species 170 Ectinorus (Ectinorus) mondacai, new species 173 '\Eluminicolal calderense, new species 61 tGastrocopta akokala, new species 33 ^'Gastrocopta kintlana, new species 32 tGobiolagus Hi, new species 258 '\Helisoma thangidcita, new species 65 tHolospira beardi, new species 41 ^'Holospira tabniini, new species . 39 Leptotarsus (Tcmypremna) gucideloupensis, new species 240 Leucania educata, new species 208 Leucania lamisma, new species 204 Leucania lobrega, new species 189 Leucania neiba, new species 217 Leucania rawlinsi, new species 186 Leucania sonroja, new species 191 Leucania toddi, new species 208 '\LucideIla Columbiana, new species 22 tLucidella salishora, new species 23 '\Lyninaea bowmanana, new species 77 tLynmaea lacerta, new species 79 tMenetus hilli, new species 70 '\Menetus textiis, new species ............................................ 69 '\Omalonyx cocleare, new species 27 Oreo helix dawsonae, new species 50 '\Oxylomal kintlana, new species 24 iPlanorbella fordensis, new species 66 '\Praticolella lucifera, new species 48 fProkanisamys major, new species 149 '(Pupoides CO status, new species 28 2001 Index to Volume 70 271 Radiocentrum kintlana, new species . 51 tSkinnerelix rothi, new species 56 ^Sphaerium discus, new species 81 fTozerpina lends, new species 19 '\Tozerpma buttsi parva, new subspecies .................................... 15 '\Tryonia russelli, new species 60 ^Vallonia kootenayorum, new species 37 •fValvata procera spatiosa, new subspecies 64 ^Vertigo consteniusi, new species 35 '\Vertigo doliara, new species 36 '\Vertigo micra, new species 36 ^Xerarionta constenii, new species 56 t Fossil taxa AUTHOR INDEX Adams, Morton S 179 Atabey, E§ref 247 Berman, David S 113 Bruijn, Hans de 143 Constenius, Kurt N 1 Dawson, Mary R 221, 257 Espinosa de los Monteros, Alejandro 231 Fonseca Zamora, Oscar 237, 263 Goehring, W. Orr 233 Hastriter, Michael W. 169 Holland, William J 235 Huang, Xue-shi 257 Pierce, Harold G 1 Reisz, Robert R. 1 13, 133 Sarag, Gergek 247 Sutherland, Tammy E 133 tinay, Engin 247 Wang, Banyue . 221 Watters, David R. 237, 263 Wessels, Wilma 143 Young, Chen W. 239 Zhang, Zhao-qun 257 .; ^ 1: ■MJ' • 'W* '■ T'^' , V- V''' ■ ^ '^ ‘ &.■ ' '■ . E'' it'WM ■ ■ y'^''^^*JH[te5L ' ' B* '*.' ''1*1 ' '* :’i: ,|i' ., , ;-',\ ^ ^ 'v ;%Ji' ■•*■ . «s ' ■■' ' 'V, % "'4?« »'«»«, :?» ' i *t&' ■' ^ *laS *)V' ^ i. j,. ■ •f''^: ,..-r-‘S',' " K,j4;S;.; 1 v%5 ^ r#J: .v J , ' '•' N>.;; '»• -;,l- ■ ‘:-fii|5Sj| ‘ 7 ; ^ ■f‘'n^i>L^',t . v.'fl 2.'^'!. ■ -'■ •^’ '*'’^ B -,. V .n ./ ?^E5Sw -"V'-'U.;, ■■ ■' '” ^..4. , V. , V - V’lV#, r- 7' ’ ) / INSTRUCTIONS FOR AUTHORS ANNALS OF CARNEGIE MUSEUM consists of contributions to the earth sciences (including paleontol- ogy), life sciences, and anthropology, in 30 by 46 pica format (127 by 195 mm or 5 by 7% inches). Submit all manuscripts to the Office of Scientific Publications. Au- thors should give particular attention to scientific con- tent, format, and general style for the ANNALS. Manu- scripts that do not conform to the style of the AN- NALS will be returned to the author immediately. Every manuscript will be subjected to external peer re- view. Authors should submit the names, addresses, phone and fax numbers, and e-mail addresses of at least four qualified potential reviewers for each manuscript. Authors will be asked to subsidize, if fimds are available, any or all costs of publication (approximately $ 100/page printed). A detailed set of instructions to authors may be requested from the Office of Scientific Publications. Manuscript Style. — Articles should include the fol- lowing items in this order: title page, abstract, text (with desired headings), acknowledgments, literature cited, ta- bles, figure captions, and copies of illustrations. All manuscripts must be typed double-spaced on standard 814- by 1 1-inch white bond paper, with at least one-inch margins all around, and submitted in quadruplicate — an original for the editors, and three review copies. All pages should be numbered including tables, literature cited and the list of figure captions. Only correspon- dence quality or better dot matrix printouts will be ac- cepted; draft mode is unacceptable. A monospaced font, (e.g.. Courier) is required. A disk in WordPerfect x.x or ASCII is helpful. Title Page. The title should be brief, include the ani- mal or plant group involved be placed on a separate page, and appear two to three inches below the top mar- gin. Include the author’s name(s) and the affiliations of non-CMNH authors. In the case of multiple authorship, indicate the address to which proofs should be sent. Abstracts. Abstracts should be short, but substantive, and included at the head of the first page of text. Text. Do not right justify text or break (hyphenate) a word at the end of a line. Footnotes and acknowledg- ments as footnotes are unacceptable. All text is double- spaced in a monospaced font (e.g.. Courier). Tables and Figure Legends. The list of figure legends and each table should be typed consecutively on indi- vidual pages separate from the text. Tables must be dou- ble-spaced throughout with no vertical lines. Measurements. Metric units should be used, if possi- ble. Literature Cited. All references to literature in text must appear in the Literature Cited section. The data (author, date, and page) in both citations must agree. Do not abbreviate the titles of periodicals or serials. The following style, double-spaced, should be used in Lit- erature Cited: 1) Two authors in a journal series: Soltis, D. E., and P. S. Soltis. 1992. The distri- bution of selfing rates in homosporous ferns. American Journal of Botany, 79:97-100. 2) Same authors repeated — use three-em dash: . 1923. The fauna of the Ardyn Obo For- mation. American Museum of Natural History Novitates, 98:1-5. 3) Same authors plus a third author — repeat all authors: Knutson, L. V, R. E. Orth, and W. L. Murphy. 1986. Catalog of Sciomyzidae (Diptera) of America north of Mexico. Entomography, 4: 1-53. 4) Chapter in an edited volume: Rausch, R. L. 1963. A review of the distribution of Holarctic mammals. Pp. 29^3, in Pacific Basin Biogeography (J. L. Gressitt, ed.). Bishop Museum Press, Honolulu, Hawaii. 5) Unpublished dissertation: Smith, J. P. 1976. Review of Eocene Mammals. Unpublished Ph.D. Dissert., University of California, Berkeley, California. 6) Book: White, M. J. D. 1961. The Chromosomes. Meth- euen and Co., Ltd., London, United Kingdom. 7) Journal articles with usual volume and issue number: Anderson, W. I. 1969. Lower Mississippian con- odonts from northern Iowa. Journal of Pale- ontology, 43:916-928. Figures. Including all illustrative materials (line art, halftones, photographs), figures are to be numbered in Arabic numerals. Four sets of figures are required, one (original artwork) for reproduction, three for reviewers. Photocopies of photographs for reviewers are usually not acceptable but are adequate for line drawing review cop- ies. Figures may not be larger than 17 by 12 inches. Reducing figures is the responsibility of the author. All figures must be reducible to a maximum of 127 by 195 mm (30 by 46 picas) without loss of clarity. Line copy should be designed for reduction to % or V2 or actual size. Typewritten figure copy will not be accepted. Pho- tographic figures should be submitted at actual repro- duction size, if possible. Rectangular halftone figures should be abutted, with- out intervening spaces. The printer will insert narrow white spaces during the reproduction process. All figures must have minimally one-inch borders all around. Each figure should be given a protective cover and identified on the back side. Lettering and/or a magnification scale (linear metric scale) for rectangular halftone figures should be placed directly on the photo, not in a blanlc space between pho- tos. The scale or lettering for closely cropped photos can be placed in blank areas close to the figure. Proof. — The author should answer all queried proof marks and check the entire proof copy. Return corrected page proof with the edited manuscript promptly to the Office of Scientific Publications. If an author chooses to make extensive alterations to a paper in proof stage, the author will bear the cost. Original manuscripts will not be returned unless request- ed. Figures will be returned to the author only if re- quested prior to publication. } ( Heckman bindery, INC. Bound-Tb-Please* APRIL 02 N. MANCHESTER, INDIANA 46962