Over a century has passed since 1901 when W.T. Hornaday showed a fragment of a horn of Triceratops found in the valley of Hell Creek to H.F. Osborn at the American Museum of Natural History. The following year Osborn's assistant, Barnum Brown, was dispatched to eastern Montana and began investigations of its geology and paleontology. By 1929, Brown had published a geological analysis of the rocks exposed in the southern tributaries of the Missouri River, named the Hell Creek Formation, and published studies of some of the dinosaurs discovered there. Parts of his collections of fossil mollusks, plants, and vertebrates contributed to research by others, particularly members of the U.S. Geological Survey. From 1930 to 1959, fieldwork was slowed by the Great Depression and World War II, but both the continuing search for coal, oil, and gas as well as collections of fossils made during construction of Fort Peck Dam set the stage for later research. Field parties from several museums collected dinosaurian skeletons in the area between 1960 and 1971. In 1962, concentrations of microvertebrates were rediscovered in McCone County by field parties from the University of Minnesota. Ten years later, field parties from the University of California Museum of Paleontology began collecting microvertebrates from exposures in the valley of Hell Creek and its tributaries. The research based on this field research provided detailed geological and paleontological analyses of the Hell Creek Formation and its biota. In turn, these contributed to studies of evolutionary patterns and the processes that produced the changes in the terrestrial biota across the Cretaceous-Paleogene boundary.

In 1907, Barnum Brown named the Hell Creek beds (Formation) for the strata exposed in the Hell Creek Valley and other downstream tributaries of the Missouri River. In the absence of a stratotype section, a lectostratotype is herein proposed for the Hell Creek Formation based on 84.2-m-thick exposures at Flag Butte (local name) in Ried Coulee (archaic use; East Fork of Hell Creek) and East Ried Coulee, tributaries of Hell Creek, Garfield County, Montana (sec. 29, T. 21 N., R. 38 W., NAD27 CONUS; base 47.55931°N, 106.88111°W; top 47.55533°N, 106.86810°W). The formation is underlain with general conformity by sandstone beds of the Fox Hills Formation (as characteristically known, the Colgate Member is absent) and is for the most part conformably overlain locally by the Tullock Member of the Fort Union Formation. The upper contact at Flag Butte is demarcated at the base of the IrZ lignite bed (above an iridium anomaly). The boundary has been demonstrated to be somewhat unconformable in areas to the west. The IrZ bed is also missing at Bug Creek in McCone County. In its type section, the Hell Creek Formation is subdivided (simply and informally) into Ried Coulee (lower Hell Creek), East Ried Coulee (middle Hell Creek), and Flag Butte (upper Hell Creek) units, each containing a sandstone and a mudstone lithofacies. Formational thickness varies with local depositional and erosional history of various coastal-deltaic environments across the Williston Basin and a trend of overall thinning to the east and northeast.

The interval spanning the uppermost Hell Creek Formation to the overlying lowermost Fort Union Formation in north-central Montana encompasses a marked paleoenvironmental change (associated with the formational contact), the Chicxulub impact event, and the Cretaceous-Paleogene boundary. We have examined the record of this transition at the Hell Creek Formation lectostratotype to determine the placement of these events using a series of lithological, geochemical, palynological, and 40 Ar/ 39 Ar geochronological analyses. The claystone derived from the Chicxulub impact is identified based on lithological criteria, enrichment of iridium and osmium, and osmium isotope ratios. The impact claystone also contains a Cyathidites fern spike. The first continuous lignite horizon in the section immediately overlies this claystone and represents the Hell Creek–Fort Union formational contact. A tuff ~3 m above the impact layer is dated to 66.024 ± 0.059 Ma. Given this evidence, at the lectostratotype the Cretaceous-Paleogene boundary is coincident with the impact claystone and therefore with the formational contact. Due to poor preservation and apparent reworking of palynomorphs surrounding the formational contact, the Cretaceous-Paleogene boundary is difficult to identify based on biostratigraphically significant taxa. The presence of marine dinoflagellates is suggestive of reworking of older marine sediments during the deposition of the Cretaceous-Paleogene boundary interval.

Magnetostratigraphic evaluation of a well-exposed stratigraphic section in northeastern Montana has been undertaken to expand upon and better understand the timing of the Hell Creek and Fort Union Formations and the biological processes recorded within them. Characteristic remanent magnetizations show clear magnetostratigraphic patterning of chrons C28n, C28r, C29n, C29r, C30n, and possibly C30r. Differentially corrected global positioning system coordinates, including elevation, were recorded at each sample site, allowing the magnetostratigraphic framework to be precisely relocated in the field and traced laterally across the landscape. In this way, important chronologic boundaries such as the Cretaceous-Paleogene boundary can be projected into or above topography of the surrounding region. Localities in Montana that have been sampled for fossil studies were mapped and correlated to the same stratigraphic sections as the magnetostratigraphy, and so they can be compared directly to the geomagnetic polarity time scale. The new magnetostratigraphy can also be used to relate to other basins of Cretaceous and Paleogene age using information independent from biostratigraphic zonation, making it possible to directly compare the composition of coeval faunas from significantly different latitudes.

Questions of biotic and environmental change during deposition of the Upper Maastrichtian Hell Creek Formation require a robust and replicable system for intra-formational correlation of fossil localities. In this paper, we present a carbon isotope chemostratigraphic curve based on terrestrial organic carbon. Data were taken from a complete measured section spanning the full 93 m of the Hell Creek Formation at our study site. Sedimentary beds were described at the centimeter scale, and samples for carbon isotope analysis were taken at ~10 cm intervals. Each sedimentary bed was analyzed in thin section, and grain-size data were assembled based on petrographic point counts. The well-documented Cretaceous-Paleogene boundary negative carbon isotope excursion, six negative carbon isotope excursions, and four tentative positive carbon isotope excursions provide chronostratigraphic tie points within the Hell Creek Formation. We used this curve to precisely correlate 12 additional fossil-bearing localities from throughout the Hell Creek Formation across its type area. These correlations revealed significant local variation in sediment accumulation rates, confirming that simple stratigraphic position relative to the diachronous base and top of the Hell Creek Formation introduces significant error in correlation.

The Hell Creek Formation in eastern Montana has yielded well-preserved leaf megafossil localities that provide insight into the vegetation and climate of the latest Cretaceous. Among the most basal, the PDM locality (UCMP [University of California Museum of Paleontology] PB99057 = MOR [Museum of the Rockies] HC-278) occurs in channel sandstones ~10 m above the underlying Fox Hills Formation. The locality represents a fluvial/estuarine environment. Leaf megafossil impressions were preserved on clay drapes within the channel. Angiosperms dominated the flora (13 of 17 morphotypes). Dryophyllum subfalcatum and “ Vitis ” stantoni , two common morphospecies in the Hell Creek Formation, are well represented. Gymnosperms including Metasequoia , Glyptostrobus , Cupressinocladus , and Ginkgo are rare; ferns and cycadophytes are absent. Univariate leaf-margin analysis produced mean annual temperature (MAT) estimates of ~7–11 °C (5–14 °C including overlapping estimation errors). The Climate Leaf Analysis Multivariate Program (CLAMP) produced a MAT value of 11–12 °C ± 2 °C. Leaf area analysis produced mean annual precipitation (MAP) estimates of 197 cm (+152/−86 cm) and 191 cm (+161/−87 cm), while CLAMP produced a growing season precipitation estimate of 82–90 ± 48 cm. The wetter MAP values are consistent with paleosols near the base of the formation, which lack paleosol carbonate. CLAMP results further suggest seasonality in both temperature and precipitation. Some PDM morphotypes are familiar from the Hell Creek I floral zone of North Dakota, and several are not, suggesting greater spatial and/or temporal heterogeneity in the Hell Creek Formation flora than has been previously appreciated.

Many workers consider the Cretaceous-Paleogene extinction the archetypal catastrophic pulse event caused solely by the Chicxulub bolide impact. However, based on a global scale analysis of marine animals, the Cretaceous-Paleogene boundary is a candidate for an extinction enhanced by the coincidence of press and pulse disturbances. We make a preliminary test of key predictions of the press-pulse hypothesis using palynological data. We document a local palynological extinction of 21% at the Cretaceous-Paleogene boundary, which is consistent with extinction rates of 15% to 30% at other localities in the Hell Creek type area and throughout North America. We also find a decline in the number of dicot angiosperm pollen taxa between −3.5 m and −2.5 m below the boundary. We document a low-palynospecies-richness interval between −1.4 m and −1.0 m that includes extirpation, but not extinction, of some palynospecies. These changes in species richness are not correlated with changes in depositional style or pollen preservation, indicating that they may represent a biological rather than entirely taphonomic signal. Review and reanalysis of previously published data from other localities in the western interior of North America suggest similar declines in species richness within approximately the same stratigraphic interval. However, many of the species absent during the low-species-richness interval reappeared before the Cretaceous-Paleogene boundary, suggesting changes in community structure and composition before the terminal Cretaceous event—a key prediction of the press-pulse hypothesis—rather than gradual extinction in the latest Cretaceous.

An examination of freshwater euselachian fossils from the Maastrichtian lower and upper Hell Creek Formation, and the Bug Creek Anthills (Cretaceous-Paleogene boundary), and the early Paleocene (Puercan) Tullock Member of the Fort Union Formation of Montana, USA, revealed seven taxa: Lonchidion selachos , Protoginglymostoma estesi , Chiloscyllium , Restesia americana , Ischyrhiza avonicola , Myledaphus pustulosus n. sp., and Dasyatis (Dasyatidae). Squatirhina americana is redescribed to the new genus Restesia based on unique crown morphology. Dasyatis is only reported from the Paleocene. This is in contrast to the other collected taxa, which are only known from the Cretaceous. Ischyrhiza is not reported in our samples from the Hell Creek Formation despite earlier erroneous claims; however, the taxon is present in a sample from the Bug Creek Anthills. We suggest that this taxon infrequently moved into the freshwater rivers to forage. Lonchidion selachos occurs only in the upper Hell Creek Formation, and we hypothesize that the upper Hell Creek localities were deposited during a warm interval, as the paleodistribution of L. selachos shifted significantly to the north during warmer periods in the Santonian and Campanian. Of the taxa found in the Hell Creek Formation and the Bug Creek Anthills, only Myledaphus is found in Paleogene deposits; however, specimens are extremely rare. The occurrence of Myledaphus in the Paleogene is suggested to be due to the occurrence of reworked material.

Remains from vertebrate microfossil localities of the Hell Creek Formation provide a basis for evaluating the diversity and relative abundance of actinopterygians. Taxonomic composition of the basal actinopterygians of the Hell Creek Formation is increased by the recognition of a probable semionotiform designated Holostean A and a small-bodied amiid. A combined taxonomic/parataxonomic approach incorporating data from centra and tooth-bearing elements demonstrates that, at minimum, 17 kinds of teleosts are present in the Hell Creek assemblage. These include an ostariophysan with possible affinities to catfish and the first Cretaceous record of Priscacara . Latitudinal patterns are identified using both presence/absence and relative abundance data. Melvius is interpreted as having a more southern distribution, while the probable semionotiform referred to as Holostean A has a more northern distribution. A greater abundance of Holostean A in the lower Hell Creek localities than in the upper Hell Creek localities is interpreted as a result of shifts in distribution in response to changes in climate. A comparison of the relative abundances of fish in assemblages from the mid-Campanian Judith River Group and the Hell Creek Formation allows changes in the structure of aquatic paleocommunities to be identified. Amiids are of much greater abundance and the osteoglossomorph Coriops is of lower abundance in the Hell Creek Formation than they are in the mid-Campanian localities. In the Hell Creek Formation, teleosts are dominated by small-bodied taxa, but greater numbers of large taxa are present, suggesting that more teleosts occupied positions higher in the food web.

Modern amphibians (lissamphibians) are highly sensitive indicators of environmental disturbance. As such, fossil lissamphibians are an excellent model for testing causal hypotheses of the Cretaceous-Paleogene mass extinction and secondary effects of Deccan volcanism and a bolide impact (e.g., acid rain). We quantitatively analyzed high-resolution temporal changes in diversity and community structure of a succession of salamander and salamander-like lissamphibian assemblages from the Hell Creek Formation and Tullock Member of the Fort Union Formation of Garfield County, northeastern Montana (ca. 67.5–65.3 Ma). Richness, evenness, and taxonomic composition remained stable through the lower Hell Creek Formation. Peak richness (11 species) occurred in the middle of the formation coincident with a short-term drop in evenness. Following a return to preexisting levels of evenness, diversity progressively declined in the upper third of the formation. This pattern reflects plummeting relative abundances of Scapherpeton tectum and a stepwise disappearance of five species, of which three represent extirpation (33%) and two represent extinction (22%). These results suggest that ecological instability increased in the local fauna during the last ~400 k.y. of the Cretaceous. Temporal correlation with local, regional, and global changes in other aspects of the terrestrial (mammals, plants) and marine (planktonic foraminifera, mollusks) biota and environment (volcanism, paleotemperature) implies a global phenomenon (late Maastrichtian event). The post–Cretaceous-Paleogene “survival” fauna from the lowermost Tullock Member was taxonomically depauperate and predominated by the “bloom taxon” Opisthotriton kayi . Together, our results lend growing support in favor of a complex multiple-cause scenario for the Cretaceous-Paleogene mass extinction event.

Climate change has been suggested to be a factor in the distribution of reptiles in the Cretaceous and Paleogene, but comparatively few studies have examined the possible role of climate within lithostratigraphically or biostratigraphically constrained local areas. Here, we reexamine the role of climatic change on the distribution of turtles within the Late Cretaceous–earliest Paleogene record of the Hell Creek area of northeastern Montana, where long-term field studies have produced a relatively dense record of faunal change during a period of climatic fluctuation. An improved stratigraphic record allows us to document previously unavailable range data for the Hell Creek–Tullock turtle taxa and assess changes in richness and abundance through approximately the last 1.9 m.y. of the Cretaceous and the first 750 k.y. of the Paleocene and provide revised estimates for survivorship and extinction across the Cretaceous-Paleogene boundary.

Here we provide a survey of Triceratops localities and accompanying stratigraphic data from the Hell Creek Formation of northeastern Montana. The majority of the sites discussed here were relocated or discovered during the last 5 yr of the Hell Creek Project (1999–2010), a multi-institutional effort to record a large volume of faunal, floral, and geologic data on the Hell Creek Formation in order to test evolutionary, paleoecological, and geological hypotheses. Triceratops is the most abundant dinosaur in the Hell Creek Formation and one of the most common nonavian dinosaurs of the Upper Cretaceous. It is known from hundreds of specimens, which have been collected since it was first described in 1889. Although these specimens provide a wealth of morphological data on Triceratops , many lack detailed stratigraphic information and context. Detailed stratigraphic and contextual data for more than 70 specimens of Triceratops collected during the Hell Creek Project make this data set among the most comprehensive for any nonavian dinosaur.

Here, we describe a juvenile Triceratops sp. skull, UCMP 136306, from the Hell Creek Formation, McCone County, Montana. The relative completeness and superb preservation of this skull contribute to an improved understanding of the cranial ontogeny, morphology, and individual variation in Triceratops . Total skull length is 120 cm long (est.). UCMP 136306 is one of the most complete Triceratops skulls of this ontogenetic stage yet known. The cranial sutures are patent, and most are overlapping with minimal sinuosity, modest interdigitation, and overlapping flat sutural surfaces. The following cranial elements are preserved and described in this study: epinasal, rostral, quadrate, quadratojugal, jugal, pterygoid, dentary, surangular, postorbital horn, parietal, squamosal, epiparietal, episquamosal, occipital condyle, supraoccipital, and exoccipital. For decades following the initial description of Triceratops by O.C. Marsh in 1889, the typical collector attitude was “bigger is better.” Emerging scientific institutions and museums with newly constructed exhibit halls demanded the biggest and newest dinosaurs. We hypothesize that this historical practice, influenced by facies and taphonomic factors in the Upper Cretaceous Hell Creek Formation, Montana (and contemporaneous formations in neighboring states), resulted in the underrepresentation of nonadult Triceratops in museum collections. This practice contributed to the false notion that nonadult Triceratops specimens are rarely preserved in the fossil record, until now.

Ceratopsid dinosaurs are notable for their common occurrences in bonebeds; however, until recently, these have not been encountered for the chasmosaurine Triceratops . The aim of this investigation is to describe the taphonomy of Quittin' Time (Museum of the Rockies locality HC-430), a Triceratops bonebed in the Hell Creek Formation, Garfield County, Montana. Using site taphonomic descriptions with an evaluation of ontogeny, inferences regarding the paleobiology of this extinct taxon are possible. The locality is associated with abundant organic material, including woody debris, large seeds, and other fragments in isolated silty lenses, all incorporated within a siltstone matrix, indicating preservation within a floodplain environment. Based on the repetition and ontogenetic stages of cranial elements, the minimum number of individuals (MNI) is three. Evidence from the location and taphonomic condition of the bones preserved in close proximity within the same siltstone unit suggests that the individuals—one young adult, one subadult, and a juvenile—likely accumulated during distinct flooding events within a narrow region of the floodplain as a result of “bloat-and-float” transport. The relatively small scale of the bonebed, both in terms of total area and number of individuals, implies that future work on Triceratops sites requires careful scrutiny of cranial elements examined within an ontogenetic framework because they are potentially critical to establishing MNI. Preservation of multiple individuals within the same unit does not necessarily provide evidence of gregarious behavior in Triceratops but rather may be a reflection of site taphonomic history and accumulation processes.

The Cretaceous-Paleogene boundary marks a critical event in mammalian evolution. Using a database of 4769 mammalian specimens from the Hell Creek and lower Fort Union formations of Garfield County, northeastern Montana, I quantified temporal patterns of diversity and community structure to evaluate faunal dynamics during the last ~1.9 m.y. of the Cretaceous, the Cretaceous-Paleogene extinction and survival, and the first ~1.2 m.y. of the early Paleocene recovery and placental radiation. Mammalian taxonomic composition and species richness remained relatively stable for most of the last ~1.9 m.y. of the Cretaceous, but the relative abundance of metatherians and evenness of paleocommunities began declining ~500–600 k.y. before the Cretaceous-Paleogene boundary. The ecological instability implied by falling evenness may be linked to the local extinction of 75% of species at or near the Cretaceous-Paleogene boundary. The early Puercan (Pu1) survival fauna from the first ~100–200 k.y. of the Paleocene is species poor and consists of a few local “bloom taxa” and an influx of immigrants. The species-rich mid/late Puercan (Pu2/3) fauna implies mammalian recovery occurred within ~600–700 k.y. of the Cretaceous-Paleogene event. The Pu2/3 and early Torrejonian (To1) faunas signal the beginning of the placental radiation—“bloom taxa” and multituberculates waned, richness and abundance of “archaic ungulates” and plesiadapiform primates increased, and the first appearance of taeniodonts, pantodonts, and possibly creodonts expanded the ecological diversity of local faunas. Together, the decoupled patterns of richness and relative abundances reveal the complexity of faunal dynamics during this seminal episode in mammalian history.