ABSTRACT: The Smoky Hill Chalk Member of the Niobrara Chalk was deposited in the Western Interior Sea over a span of about five million years extending from the upper Coniacian through the lower Campanian. Coincidentally, this period also encompasses much of the early evolution and radiation of the family Mosasauridae. Thousands of mosasaur specimens have been collected from the Smoky Hill Chalk Member of western Kansas since the first Yale College Scientific Expedition in 1870. Early workers viewed the entire Smoky Hill Chalk Member as the "Pteranodon beds" and were not able to provide accurate stratigraphic information for the material they collected. Even though mosasaurs are represented by numerous complete and well preserved specimens, the ranges of the various species could not be defined further without better stratigraphic data. In 1898, S. W. Williston first described the general distribution of mosasaurs within the Rudistes and Hesperornis beds of the Smoky Hill Chalk Member. Seventy years later, D. A. Russell grouped mosasaur species into upper and lower zones in the chalk and in 1990, J. D. Stewart incorporated D. E. Hattin's stratigraphic data into his biostratigraphic zones and refined the occurrences of mosasaur species to a higher degree than had been done previously. B. A. Schumacher in 1993, and M. A. Sheldon in 1996, reviewed existing collections and provided further definition to mosasaur biostratigraphy. Recently reported specimens of Tylosaurus proriger, Platecarpus planifrons and Clidastes liodontus with accurate stratigraphic information have further expanded and refined the temporal distribution of these species within the Smoky Hill Chalk Member.

Introduction

Although the recovery of mosasaur remains (Mosasaurus hoffmanni) from a limestone mine in the Netherlands about 1780 pre-dates the Mantell's (1842, p. 61) term ‘Dinosauria’ by more than half a century, there is still much to be learned about this highly successful group of marine reptiles. In the last 200 years, thousands of mosasaur specimens have been collected from around the world. The majority of these remains have been discovered in marine sediments that accumulated in the Western Interior Sea of North America during Late Cretaceous time. By far the largest number, however, have come from the Smoky Hill Chalk Member of the Niobrara Chalk in western Kansas.

Harlan (1834) wrote: "It is not improbable that Lewis and Clarke, in their Expedition up the Missouri, allude to the remains of a similar animal in the following extracts. "Monday, September 10th, 1804, we reached an island (not far from the grand detour, between Shannon creek and Poncarrar River), extending for two miles in the middle of the river, covered with red cedar, from which it takes the name of Cedar Island; just below this island, on a hill, to the south, is the back-bone of a fish forty-five feet long, tapering towards the tail, and in a perfect state of petrifaction, fragments of which were collected and sent to Washington." Although the specimen was subsequently lost, it is more likely that it was the remains of a large mosasaur (Martin and others, 1998) than a fish. The German naturalist, G. A. Goldfuss (1845) documented the occurrence of Mosasaurus in the United States from remains discovered along the Missouri River in South Dakota. Based upon the limited information available to him at the time, Owen (1851) stated that the report of the mosasaur discovery in North America by Goldfuss provided "a highly instructive instance of the coexistence of particular forms of fossil Reptilia in remote parts of the earth, at the same geologic epoch." In 1868, the first specimen of Tylosaurus proriger (Cope 1869; MCZ 4374) was collected by "Colonel Conyngham and Mr. Minor" from near Monument Rocks  in western Kansas (Williston, 1898), and the region quickly became recognized as a productive source of mosasaur and other marine reptile remains.

The Yale College Scientific Expeditions to Kansas by O. C. Marsh in the early 1870s collected more than 2000 mosasaur specimens from exposures that would later become known as the Smoky Hill Chalk Member. At that time, the first transcontinental railway had just been completed across Kansas, and the western part of the State was only sparsely settled. Eastern paleontologists, such as Marsh, usually carried guns and required Army escorts for protection, but Williston (1898) wrote that he and others went largely unarmed during their explorations. Although credit is usually given to Cope and Marsh for their early work with Cretaceous fossils from Kansas, others, including B. F. Mudge, Dr. George F. Sternberg, Dr. Theophilus Turner, and Dr. John Janeway  were the real pioneers of paleontology in the State (Williston, 1898). Mudge's work  is recognized by Cope (1872a; see also this note), and many of Dr. Sternberg's mosasaur specimens are in the United States National Museum (Smithsonian). At the time, the geology of Kansas was largely unknown and consequently there was little or no stratigraphic information associated with this wealth of well-preserved material. The intense scientific interest regarding these marine reptiles was brief, however, as the lure of dinosaur remains farther west in Colorado and Wyoming soon attracted the attention of Cope and Marsh. Consequently, major institutional collecting efforts in the Niobrara of Kansas essentially ceased for the next 100 years.  The work of Charles H. Sternberg and his family is a notable exception.

According to Zakrzewski (1996), geologic studies of the Western Interior of the United States had begun as early as the 1850s, but had proceeded slowly and sporadically. Meek and Hayden (1861) first referred to the chalk and limestone strata as the Niobrara Division in their description of exposures along the Missouri River near the mouth of the Niobrara River in Nebraska. In Kansas, these Upper Cretaceous strata were referred to simply as the 'Niobrara' by the geologists and paleontologists of the day (Hattin, 1982).

E. D. Cope (1872b) wrote the earliest substantial account of vertebrate fossils from the "Niobrara Beds." According to Cope (1872b), the "study of this chapter of ancient history" was only five years old and there were already 17 species of mosasaurs known from the Cretaceous of Kansas. In describing the geology of the chalk and each of the various species of marine reptiles in some detail, he made no attempt to delineate their stratigraphic occurrence within the formation.

In 1889, the University Geological Survey of Kansas was established by the state legislature. Samuel W. Williston was appointed to the faculty of the University of Kansas the following year. These two events provided the basis for much of the early progress that was to be made in the study of geology and paleontology in Kansas. Even reaching an agreement on what to name the formations that cropped out in the western part of the state was no easy matter. Most of the early terminology used to describe the Smoky Hill Chalk Member was based on the occurrence of the predominant fossils and added little stratigraphic information to individual specimens. The "Niobrara Division" of Meek and Hayden (1861) was made up of two distinct units; a lower limestone member and an upper chalk member. Logan (1897) first termed the chalk member "The Pteranodon Beds" in apparent recognition of the abundance of well-preserved Pteranodon material that had been discovered therein. That same year, demonstrating his support for Logan's descriptive terminology, Williston (1897) further divided the Pteranodon Beds into the lower Rudistes Beds and the upper Hesperornis Beds, providing essentially the first biostratigraphic subdivisions of what was to become the Smoky Hill Chalk Member.

Williston (1897) briefly discussed the stratigraphic occurrence of mosasaurs in the Pteranodon Beds for the first time. He also made the observation that Clidastes does not occur in the Rudistes Beds, indicating that other genera probably occurred within 100 feet of the contact of the chalk with the underlying Fort Hays Limestone. Williston (1897) was certainly aware of the lack of good stratigraphic data for Niobrara vertebrate fossils when he wrote, "I need not call the attention of future collectors to the importance of locating the horizon of specimens more accurately than has been done heretofore."

In the years that followed, S. W. Williston, Charles H. Sternberg and sons, H. T. Martin, and others continued to collect spectacular examples of mosasaurs and other marine species from the Smoky Hill Chalk Member without adding significantly to the knowledge of the stratigraphic record of these creatures. In 1898, Williston published the first comprehensive description of the systematics and comparative anatomy of mosasaurs from the Smoky Hill Chalk Member, and discussed their range and distribution in comparison with specimens discovered earlier in New Zealand and Europe. He commented that Tylosaurus, "so far as was known, begins near the lower part of the Niobrara [Smoky Hill Chalk Member] and terminates at its close or in the beginning of the Fort Pierre [Pierre Shale]." Of Platecarpus, he stated that the species on which the genus is based are "known nowhere outside of Kansas and Colorado, and are here restricted exclusively to the Niobrara." He again concluded that the lowest horizon of Clidastes "is the upper part of the Niobrara in Kansas." It was not until after the turn of the century that the exploration for oil and gas in western Kansas enabled rapid advances in understanding the geology of the entire Niobrara Formation. According to Hattin (1982), Moore and Hays (1917) were the first to regard the Kansas Niobrara as a formation, and the first to give member status to the currently recognized divisions; the Fort Hays Limestone Member and the Smoky Hill Chalk Member.

Russell (1967) reviewed the specimens collected by the 1871 Yale expedition and postulated that the Smoky Hill Chalk Member could be divided into a lower, Clidastes liodontus-Platecarpus coryphaeus-Tylosaurus nepaeolicus zone and an upper, Clidastes propython-Platecarpus ictericus-Tylosaurus proriger zone (Figure1). He also suggested that the increased abundance of Clidastes specimens in the upper portion of the chalk was an indication of a gradual change from a mid-ocean to a near-shore environment. Russell (1970) noted significant differences between the distribution of mosasaur species in the Smoky Hill Chalk Member compared to the Gulf Coast species occurring in the Selma Formation of Alabama. In his initial paper concerning the biostratigraphy of the Smoky Hill Chalk Member, Stewart (1988) stated that he was aware of several exceptions to Russell's stratigraphic distribution of mosasaurs in the Smoky Hill Chalk Member that caused him to regard it with "a degree of skepticism."

It was not until Hattin (1982) published his composite measured section of the Smoky Hill Chalk Member that significant progress could be made in understanding the vertebrate biostratigraphy of this formation. Hattin used bentonites and other features to delineate his 23 lithologic marker units and divided the chalk into five biostratigraphic zones based on the occurrence of invertebrate species. In doing so, he provided field workers with the first dependable method of determining their stratigraphic location in the section.

Stewart (1990) incorporated Hattin's marker units as upper and lower boundaries for his six proposed biostratigraphic zones. He provided the first comprehensive description of the distribution of known invertebrate and vertebrate species in the Niobrara Formation and made the first attempt to assign specific stratigraphic ranges for mosasaur species within the Smoky Hill Chalk Member. Even with the substantial improvements that had been made over previous attempts, Stewart believed that his biostratigraphy was flawed by the lack of reliable stratigraphic data for even those specimens collected during the previous 20 years. He stated that his framework was "submitted in the hopes that other researchers will test it and improve upon it."

Schumacher (1993) and Sheldon (1996) reviewed existing collections of mosasaur material in the Sternberg Museum of Natural History, the Yale Peabody Museum, and other institutions, and further refined the occurrence of mosasaurs in the Smoky Hill Chalk Member by building upon the stratigraphic methodology provided by Hattin (1982) and Stewart (1990). Since that time, additional discoveries (Everhart and Everhart, 1997; Everhart, Everhart and Bourdon, 1997; and Everhart and Johnson, 2001) of mosasaur remains with associated stratigraphic data have aided in the refinement of the known temporal ranges of Tylosaurus proriger, Platecarpus planifrons and Clidastes liodontus. Additional collecting is required to improve the accuracy of these ranges, and it is essential that good stratigraphic information be obtained for all specimens from the Smoky Hill Chalk Member.

Institutional Abbreviations

AMNH, American Museum of Natural History, New York, NY; FFHM: Fick Fossil and History Museum, Oakley, KS; FHSM, Fort Hays Sternberg Museum, Hays, KS; YPM, Peabody Museum of Natural History, Yale University, New Haven, CT.

The Smoky Hill Chalk Member

The Smoky Hill Chalk Member is the uppermost member of the Niobrara Chalk (Late Cretaceous) and was deposited near the middle of the Western Interior Sea between 87 and 82 million years ago from upper Coniacian through lower Campanian time (Obradovich, 1993). The member crops out in portions of northwestern Kansas as well as parts of Colorado, Nebraska, and South Dakota (Schumacher, 1997). The chalk itself consists primarily of the compacted skeletal remains (coccolithophores and coccoliths) of planktonic, golden-brown algae (Chrysophyceae). According to Russell (1967), the present-day exposures of the chalk in western Kansas were deposited about 200 miles from the eastern shoreline of the Western Interior Sea, and some 600 miles from the western shoreline. Changing environmental conditions during Coniacian time, including the gradual narrowing and shallowing of the seaway, resulted in the transition from the underlying Fort Hays Limestone to the Smoky Hill Chalk Member. In the area of deposition in western Kansas, the depth of the sea has been estimated to have been between 150 and 300 m (Hattin, 1982). For the most part, accumulation of the chalk forming sediments occurred during the regressive phase of the Niobrara depositional cycle.

The continuous, nearly horizontal, and well-defined chalk layers that are observed across the member suggest that conditions in the shallow sea during this period were relatively calm. Hattin (1982) determined a thickness of 181.8 m (596.3 ft) for his composite section and calculated a depositional rate of approximately 0.036 mm / yr. (700 yrs / in). Currents on the sea floor were minimal and there is little or no indication of sorting of shell debris, bones, or other preserved materials. Bottom muds were firm enough to support large invertebrates and vertebrate skeletal material for a sufficient amount of time that colonization by oysters could occur. Giant inoceramid bivalves and rudists were abundant at various times. Numerous bentonite layers, which occur throughout the chalk, were derived from the widespread, aerial deposition of volcanic ash and are evidence of considerable volcanic activity to the west. Although the thickness of many of these bentonites indicates that significant volcanic eruptions had taken place, there are no known mass death assemblages associated with these events. Individual bentonites are traceable for many miles across the chalk exposures and are important to the definition of the marker units used in stratigraphic determinations. Marker Unit 10 (Hattin, 1982) forms an easily recognizable caprock near the mid-point of the Smoky Hill Chalk Member. As a general rule, the more easterly exposures (Ellis, Trego, and Gove counties) are from lower in the member (upper Coniacian to lower Santonian) than those in Wallace and Logan counties to the west (middle Santonian to lower Campanian).

As most of the collecting from the Smoky Hill Chalk Member was done many years before Hattin's (1982) work, the lack of good geological data makes the early stratigraphic determinations and localities somewhat suspect. In his History of the State of Kansas, Cutler (1883) published estimates of the approximate thicknesses of the various geologic formations in Kansas and indicated that the Niobrara (Smoky Hill Chalk Member) was only 200 ft thick. More than 25 years after most of the original collecting was done on the chalk, even Williston (1897) underestimated the thickness of the chalk by as much as 33%.

The Occurrence of Mosasaurs

Mosasaurs were relatively late arrivals in the Mesozoic seas, with a fossil record that apparently began in Kansas during lower Turonian time (Martin and Stewart, 1977). Although the exact origin of these marine reptiles is somewhat uncertain, they appear to have evolved rapidly from shore-dwelling aigialosaurs / coniasaurs (Bell, 1997), and by the upper Coniacian had become the dominant marine predator in the Western Interior Sea. Coniasaur remains have been located in Cenomanian deposits in Kansas (Liggett and others, 1997). Russell (1993) and Bakker (1993) indicated that another group of marine reptiles, the ichthyosaurs, had disappeared from North America by the end of the Cenomanian. Russell (1993) suggested that mosasaurs were able to fill many of the ecological niches left vacant by the demise of ichthyosaurs. Bakker (1993) proposed that plesiosaurs replaced the ichthyosaurs and that mosasaurs may have replaced the marine crocodilians that had disappeared by the beginning of the Coniacian in the Western Interior Sea. Because specimens of mosasaurs outnumber those of plesiosaurs by at least 10 to one in the Niobrara and Pierre formations (Russell, 1967), it appears that they were able to successfully compete with the plesiosaurs and eventually dominated the Western Interior Sea during the Late Cretaceous. A Tylosaurus specimen discovered by C. H. Sternberg (1922) included a juvenile polycotylid plesiosaur (Dolichorhynchops osborni) as stomach contents, indicating that a predator-prey relationship between the two groups of marine reptiles (see also, Everhart, 2003).

Although mosasaur remains have been discovered in Late Cretaceous marine deposits from every continent, including Antarctica (Novas. et al., 2002), the sediments laid down in the Western Interior Sea, especially the Smoky Hill Chalk Member, have consistently produced the largest number of well-preserved and nearly complete specimens. According to Williston (1898), "Kansas, par excellence, has been the great collecting ground of the world for these reptiles." The majority of the mosasaur remains in the Smoky Hill Chalk Member represent three genera: Tylosaurus, Platecarpus, and Clidastes. Other genera are rare, and most are represented by single or incomplete specimens (Russell, 1967; Lindgren and Everhart, 2000).

Based on the collections made by the Yale College Scientific Expeditions in the early 1870s, the accepted view has been that certain species occur only in the lower chalk (Rudistes Beds) and others occur only in the upper chalk (Hesperornis Beds). This idea was proposed first by Williston (1897) and has been generally repeated by workers since that time (Figure 2). Russell (1967, 1988, 1993) suggested that Tylosaurus nepaeolicus, Platecarpus coryphaeus, and Clidastes liodontus were lower chalk species and that Tylosaurus proriger, Platecarpus ictericus, and Clidastes propython were upper chalk species (Figure 3) without providing definition of the stratigraphic boundaries of these units. Bennett (2000) reported that the upper half of the Smoky Hill Chalk Member has produced the majority of fossil vertebrate specimens. This suggests that the accepted distribution of mosasaur species (Russell, 1967, Schumacher 1993, and Sheldon, 1996) may reflect a significant collecting bias.

There are about 2000 mosasaur specimens from the Smoky Hill Chalk Member of Kansas in the Yale Peabody Museum that were collected by the Yale College Scientific Expeditions of the early 1870s. Only a few hundred of these are identified to at least the generic level. Any analysis of the stratigraphic distribution of these early specimens is compromised by the lack of reliable locality data. Previous authors have used data from this large and unique collection to draw conclusions regarding the relative abundance, occurrence, and other relationships between the various genera of mosasaurs, as well as the ecological significance of these genera. Most of the mosasaur material currently available for study was collected during the 1870s before the geology of Kansas was well understood and before much of the western portion of the state had been accurately surveyed and mapped. Several of the counties where many of the mosasaur remains were collected (Gove, Logan, and Wallace) by the Yale College Scientific Expeditions were not organized when William G. Cutler (1883) wrote his History of the State of Kansas. According to Bennett (2000), almost all of the YPM collection was obtained before the original Wallace County was divided into Wallace and St. John counties in 1881, and before St. John County became Logan County in 1885.

The phylogeny of mosasaurs has also undergone significant changes since that time, and is currently under review (G. L. Bell, pers. comm., 1998). In addition, a third, undescribed species of tylosaur (Stewart, 1990, Schumacher, 1993, Sheldon, 1996, and Bell, 1997) and a previously unreported species of Platecarpus (Schumacher, 1993; Sheldon, 1996; Bell; 1997; and Everhart and Johnson, 2001) recently have been added to the list of vertebrates in the Smoky Hill Chalk Member. (NOTE: Everhart (2005) described and named a new species, Tylosaurus kansasensis, from the lower Smoky Hill Chalk (upper Coniacian). This species had been previously referred to as Tylosaurus n. sp. by Stewart (1990) and Tylosaurus sp. novum by Bell, (1993; 1997). Bell (ibid.) concluded that it was the most basal of the tylosaurines.)

Besides variations in the frequency of their occurrence, these genera also differed greatly in size. It is likely that a preservational bias favored larger individuals in the Western Interior Sea. An adult Clidastes grew to a maximum length of about 4 m and Platecarpus rarely exceeded 7 m in length. Tylosaurus is by far the largest of the three genera, and probably reached lengths of 10 m or more during the deposition of the Smoky Hill Chalk Member. The adult lengths of all three genera increased through geologic time. Williston (1897) suggested that only the larger mosasaurs such as Tylosaurus ventured out into the middle of the Western Interior Sea and that the more abundant occurrence of turtles, small pterodactyls, and toothed birds in the upper Hesperornis Beds indicated a shallowing sea and an approaching shoreline. Although indicating that the nearest shoreline was 200 or more miles to the east, Russell (1967) agreed with Williston, stating that Tylosaurus and Platecarpus seemed to frequent deeper waters further from the coastlines. Both implied that the smaller Clidastes species were more likely to be near-shore dwelling mosasaurs and that Clidastes became abundant only during latter stages of the Smoky Hill deposition in response to the narrowing of the sea. On the basis of comparisons of bone microstructure, however, Sheldon (1995, 1996, 1997) proposed that Clidastes and Tylosaurus were better adapted for diving in deeper water, whereas Platecarpus was best suited for shallower coastal waters. Russell (1970) suggested that Platecarpus may have been better suited to the cooler waters in the Western Interior Sea, whereas Clidastes preferred the warmer waters of the Gulf Coast. This preference may have been related to feeding habits and food sources. Russell (1970) noted that the belemnites, which were common at more northern latitudes during the Late Cretaceous, are absent in the Mooreville Chalk of the Gulf Coast and suggested that these cephalopods may have been a significant portion of the diet of the more slender-toothed Platecarpus.

1872 Map of Kansas. Note that Wallace County (extreme left) occupies the present area of Wallace and Logan counties. The line of counties stretching westward along the Kansas Pacific Railroad (now Union Pacific) from Fort Riley to Fort Harker to Fort Hays to Fort Wallace generally provide the basis for the settlement of western Kansas.  Almost all the early towns along this stretch are 'railroad' towns, and many of those (Sheridan, Gopher, etc.) are now renamed or abandoned. The counties north and south of the railroad were not incorporated until several years later.

Locality information, which generally is provided only to the county level, seems to support Russell's (1967) contention that specimens from the eastern exposures (Gove, Ellis, and Trego counties) occur nearer the base of the formation and are older than those recovered from the western exposures (Logan, and Wallace counties). In some situations, where reliable locality information is available for a museum specimen, sites can be revisited (Schumacher, 1993, and Sheldon, 1996) to determine the biostratigraphy with a reasonable degree of accuracy. On-going and future research into the use of nannofossils identified from the matrix associated with these early collections may provide the tools necessary to determine where in the stratigraphic column the earlier specimens were preserved. Beyond these general assumptions, it is necessary that additional material associated with accurate stratigraphic information be obtained before a reasonably valid description of the evolutionary and ecological relationships between these mosasaur genera can be proposed.

Discussion of New Material

The author had the opportunity to participate in the recovery of three relatively complete and readily identifiable mosasaur specimens from the Smoky Hill Chalk Member since publication of the most recent papers on mosasaur biostratigraphy by Schumacher (1993) and Sheldon (1996). Detailed stratigraphic information was obtained during the fieldwork for each of the specimens. Two of these specimens have been reported by Everhart and Everhart (1997) and by Everhart and others (1997). A third specimen, which the author helped recover and prepare, was examined and tentatively identified by C. R. Kiernan (pers. comm., 1998) as Platecarpus planifrons through comparison with the type specimen in the American Museum of Natural History. The discovery was reported at the 2001 SVP meeting in Bozeman, MT (Everhart and Johnson, 2001).

Clidastes liodontus Merriam 1894

In July 1995, the skull and 18 vertebrae of a small Clidastes liodontus (FHSM VP-13909) were discovered eroding from a chalk exposure in southern Gove County, Kansas. Although the anterior portion of the skull had been damaged by weathering, the remainder of the specimen was recovered in situ and was generally articulated. Identification of the specimen was based on the characteristics of the premaxillary (bases of posterior teeth not emerging into the premaxillary - maxillary suture), the frontal, the parietal, and the quadrates. The remains were located immediately below a narrow, red-orange bentonite, which was identified as the base of Hattin's (1982) Marker Unit 4. This marker unit, located in the lower one fourth of the member, is latest Coniacian in age.

The biostratigraphic occurrence of Clidastes has been in question since Williston (1897, 1898) first suggested the genus was limited to the upper chalk. Additional specimens had extended that range slightly downward in the chalk, but no significant changes to the range of Clidastes have been made since Williston's time. Russell (1967, 1988, 1993) considered Clidastes liodontus to be part of a lower chalk fauna, but did not provide a definition of the term "lower chalk" in regard to its stratigraphic occurrence.

Schumacher (1993) stated that C. liodontus appeared to be absent from the lowest chalk and cites a single specimen in the Sternberg Museum collection (FHSM VP-2071), collected from Logan County, that may have come from just below the midpoint of the section, possibly as low as Marker Unit 9. According to Sheldon (1996), Clidastes does not seem to be present below Marker Unit 9 (Figure 4). However, she stated that she had located records of 18 specimens from Ellis, Gove, Trego, Rooks, and Graham counties that could have come from lower in the section. In addition, she documented records of 74 other Clidastes specimens without stratigraphic data from the Smoky Hill Chalk Member that only listed "Kansas" as the locality.

The new specimen documents the presence of Clidastes liodontus at Marker Unit 4, 35.4 m (116 ft) above the base of Hattin's (1982) composite section. The remains were recovered from within Stewart's (1990) biostratigraphic zone of Protosphyraena perniciosa. This specimen, and another specimen consisting of four dorsal vertebrae (FHSM VP-13908) from the same horizon, firmly establishes Clidastes liodontus in the upper Coniacian fauna of the Western Interior Sea.

Tylosaurus proriger (Cope 1869)

In the summer of 1996, the skull and cervical vertebrae of a large (10 m) mosasaur were discovered eroding out of the upper end of a canyon in Gove County. At the request of the landowner, the 1.2 m skull was removed and prepared by the author in 1996, and the remainder of the post-cranial material was recovered in a series of digs by volunteers in 1997. The remains were identified as Tylosaurus proriger on the basis of the large pre-dental extension of the rostrum, the location of the parietal foramen, the size and shape of the quadrates, and the anterior extension of the postorbital frontal over the posterior wing of the prefrontal. Stratigraphically, the remains were located below the base of Marker Unit 10 (Hattin, 1982) making it the earliest documented example of this species. In 1997, the specimen (FFHM 1997-10) was donated by the landowners to the Fick Fossil and History Museum in Oakley, Kansas.

Williston (1897) did not comment on the occurrence of tylosaurs except to say that the genus occurs to within 100 feet of the base of the Smoky Hill Chalk Member. Russell (1967) indicated that Tylosaurus nepaeolicus is one of the species of mosasaurs from the "lower zone" of the chalk. The range of Tylosaurus proriger, on the other hand, begins in the "upper zone" of the chalk and extends upward into the lower part of the Pierre Shale. The stratigraphic limits of the lower and upper zones were not defined by Russell (1967, 1988, 1993), and it is apparent that he and other workers were hampered by the lack of an accurate tool for such determinations.

Stewart (1990) was the first to utilize the stratigraphic marker units proposed by Hattin (1982) to delineate the occurrence of mosasaurs. He placed Tylosaurus nepaeolicus and an undescribed tylosaurid in his biostratigraphic zone of Protosphyraena perniciosa (below Marker Unit 5), and Tylosaurus proriger in the upper chalk zones of Spinaptychus sternbergi and Hesperornis (above Marker Unit 11). He also stated, somewhat prophetically, that T. proriger may occur lower in the Smoky Hill Chalk Member. Schumacher (1993) reviewed tylosaur specimens at the Sternberg Museum and concluded that Marker Unit 11 was the lower limit for T. proriger. In her review of the Yale Peabody Museum collection, Sheldon (1996) did not report specimens of T. proriger from below Marker Unit 15.

The new specimen of Tylosaurus proriger was located 2.7 m below the lowest of five bentonite seams at the base of Hattin's (1982) Marker Unit 10, or about 88.5 m (290 ft) above the base of the composite section. This stratum is near the top of Stewart's (1990) biostratigraphic zone of Clioscaphites vermiformis and C. choteauensis and is middle Santonian in age. Additional remains (Tylosaurus sp. skull; FHSM VP-13742 and Tylosaurus sp. dorsal vertebrae; FHSM VP-13908) that are attributable to an as yet unidentified species of large tylosaur have been recovered from between Marker Unit 4 and 5. Although definitive diagnostic material is lacking for these specimens, the large size of the remains suggests that the individuals were probably Tylosaurus proriger and not T. nepaeolicus (See Everhart, 2002 for updated information on the size ranges of these two species). I conclude that Tylosaurus proriger may have appeared as early as the upper Coniacian (Marker Unit 5). The recent determination of the stratigraphic location of a large T. nepaeolicus specimen (FHSM VP-7262) at Marker Unit 5 documents a provisional uppermost occurrence of this species and confirms the assertion of Schumacher (1993).

(Platecarpus planifrons Cope 1874)

Now Selmasaurus johnsoni Polcyn and Everhart 2008

In 1996, an amateur collector on a family outing in eastern Gove County located several vertebrae eroding from the base of a chalk column. He contacted the author for help in identifying the vertebrae and it was determined that the material came from a small mosasaur. The vertebrae are well preserved and relatively uncompressed. In May of 1997, the author assisted with the excavation of the remains and measured the stratigraphy. Most of the skull, cervical and dorsal vertebrae, and several ribs were recovered. All of the bones were disarticulated and scattered randomly over a two square meter area. The remains had apparently been exposed on the sea floor long enough to be scavenged, and then colonized by oysters (Pseudoperna congesta).

Although the remains initially appeared to be from a juvenile Platecarpus, there were significant differences between it and a specimen of Platecarpus tympaniticus used for comparison. The remains (FHSM VP-13910) were sent to C. R. Kiernan for further examination and identification. The specimen was compared to the type specimen of Platecarpus planifrons at the American Museum of Natural History and it was determined that it was similar, but not identical to, the type. Because the relationship between Platecarpus tympaniticus and P. planifrons is somewhat unclear at this point, final identification is reserved pending further examination. For the purpose of this paper, the specimen is attributed to Platecarpus cf. P. planifrons.

Williston (1897) suggested that Platecarpus and Tylosaurus occur throughout the Pteranodon Beds (Smoky Hill Chalk Member) but did not distinguish between the various species. According to Russell (1970), Platecarpus is the most abundant genus of mosasaur in the Niobrara Chalk, making up almost 60% of specimens in the Yale Peabody Museum collection. Russell (1967) suggested that Platecarpus ictericus was an upper chalk species and Platecarpus coryphaeus was a lower chalk species. He also stated that Platecarpus tympaniticus is probably the senior synonym for both taxa. Nicholls (1988) did not recognize the validity of P. ictericus or P. coryphaeus, and Stewart (1990) stated that the distinction between the two is questionable. Schumacher (1993) and Sheldon (1996) agree with Stewart. In his revision of mosasaur phylogeny, Bell (1997) included only P. tympaniticus and P. planifrons. On the basis of this convention, Platecarpus tympaniticus is here considered to be the senior synonym for P. ictericus and P. coryphaeus, and there is no upper-lower stratigraphic distinction. P. tympaniticus occurs from the base of the Smoky Hill Chalk Member and into the overlying Pierre Shale.

Conclusions

Within 100 years of the first discovery of Mosasaurus hoffmanni  in the Netherlands, thousands of mosasaur specimens had been collected from the Smoky Hill Chalk Member of western Kansas. The lack of good stratigraphic data for most of this material has limited the utility of determinations concerning the occurrence and ecological relationships of these marine reptiles. Recently discovered specimens of Clidastes liodontus, Tylosaurus proriger, and Platecarpus  planifrons from the Smoky Hill Chalk Member of Kansas are associated with accurate stratigraphic data. This information significantly extends the known ranges of these mosasaur species. The most recent data on the stratigraphic occurrence of mosasaurs (Schumacher, 1993 and Sheldon, 1996) is modified in Figure 5 to include these specimens. The earlier occurrence of Clidastes liodontus and Tylosaurus proriger may provide additional insights into the origin of these highly successful reptiles and adds to the faunal record of the Western Interior Sea. Additional collecting, in association with accurate stratigraphic information, is essential to future refinements to the fossil record of mosasaurs.

Acknowledgments

I thank J. D. Stewart and Donald Hattin for the assistance they have provided during the past 14 years in discussions regarding the stratigraphy of the Smoky Hill Chalk Member. Without their insights, this paper would not have been possible. I also appreciate the patient help of Gorden Bell, Bruce Schumacher, and C. R. Kiernan with the examination and the identification of mosasaur material. I thank Jerome (Pete) Bussen of Wallace, Kansas, for sharing his considerable knowledge of Niobrara stratigraphy and his enthusiasm for collecting. Mr. and Mrs. Glenn Bird of Quinter, Kansas, provided me with the 'once in a lifetime' opportunity to work on the early Tylosaurus proriger specimen. Jim and Jesse Bourdon of Croton-on-Hudson, New York, and Steve Johnson of Wichita, Kansas, generously allowed me access to their mosasaur specimens. Larry Martin and Desui Maio have been supportive in allowing me access to mosasaur material in the Natural History Museum at the University of Kansas.  I sincerely appreciate the help given by Dan Merriam, Richard Zakrzewski, Kenneth Carpenter, Greg Liggett and Earl Manning in the proofreading of this manuscript. Its accuracy and readability are due in large part to their efforts and all remaining errors are mine. Lastly, I thank my wife, Pamela, for her support, understanding and companionship in the field.

Literature Cited

Bakker, R. T. 1993. Plesiosaur extinction cycles - events that mark the beginning, middle and end of the Cretaceous. Pages 641-664 in Caldwell, W. G. E. and E. G. Kaufmann, eds., Evolution of the Western Interior Basin, Geol. Assoc. Canada, Special Paper 39.

Bell, G. L. 1997. A phylogenetic revision of North American and Adriatic mosasauroidea. Pages 293-332 in Callaway, J. M. and E. L. Nicholls, eds., Ancient Marine Reptiles, Academic Press, San Diego.

Bennett, S.C. 2000. Inferring stratigraphic position of fossil vertebrates from the Niobrara Chalk of western Kansas. Current Research in Earth Sciences, Kansas Geol. Survey Bull. 244(1): 1-26. http://www.kgs.ukans.edu/Current/2000/bennett1.html

Cope, E. D. 1872a. Note of some Cretaceous vertebrata in the State Agricultural College of Kansas. Proc. Amer. Phil. Soc. 12(87): 168-170. (for Oct. 20, 1871 meeting)

Cope, E. D. 1872b. On the geology and paleontology of the Cretaceous strata of Kansas. Part III, Paleontology. Pages 318-349 in Hayden, F. V., ed. Preliminary Report of the United States Geological Survey of Montana and Portions of Adjacent Territories, Government Printing Office.

Cutler, W. G. 1883. History of the State of Kansas. A. T. Andreas, Chicago. 1,616 pages.

Everhart, M. J. 2002. New data on cranial measurements and body length of the mosasaur, Tylosaurus nepaeolicus (Squamata; Mosasauridae), from the Niobrara Formation of western Kansas. Kansas Academy of Science, Transactions 105(1-2): 33-43.

Everhart, M.J. 2005. Earliest record of the genus Tylosaurus (Squamata; Mosasauridae) from the Fort Hays Limestone (Lower Coniacian) of western Kansas. Transactions 108 (3/4): 149-155.

Everhart, M.J. 2005. Tylosaurus kansasensis, a new species of tylosaurine (Squamata: Mosasauridae) from the Niobrara Chalk of western Kansas, U.S.A. Netherlands Journal of Geosciences / Geologie en Mijnbouw, 84(3), p. 231-240.

Everhart, M. J. and P. A. Everhart. 1997. Earliest documented occurrence of the mosasaur, Tylosaurus proriger, from the Smoky Hill Chalk (Niobrara Formation, upper Cretaceous) of western Kansas. Kansas Acad. of Sci. Trans., Abstracts, 16: 14.

Everhart, M. J., P. A. Everhart and J. Bourdon. 1997. Earliest documented occurrence of the mosasaur, Clidastes liodontus, in the Smoky Hill Chalk (Upper Cretaceous) of western Kansas. Kansas Academy of Science, Transactions, Abstracts, 16:14.

Everhart, M. J. and  S. E. Johnson. 2001. The occurrence of the mosasaur, Platecarpus planifrons, in the Smoky Hill Chalk (Upper Cretaceous) of western Kansas. Jour. Vert. Paleon. 21(suppl. to 3):48A. Abstract

Goldfuss, G. A. 1845. Der Schädelbau des Mosasaurus, durch Beschreibung einer neuen Art dieser Gattung erläutert. Nova Acta Academa Ceasar Leopoldino-Carolinae Germanicae Natura Curiosorum 21: 173-200, pls. VI-IX.

Harlan, R., 1834. Notice of the discovery of the remains of the Ichthyosaurus in Missouri. N. A.. Trans. Amer. Philos. Soc. 4: 405-409, pl.20.

Hattin, D. E. 1982. Stratigraphy and depositional environment of Smoky Hill Chalk Member, Niobrara Chalk (upper Cretaceous) of the type area, western Kansas. Kansas Geol. Survey Bull. 225, 108 pp.

Konishi, T. and Caldwell, M. W. 2007. New specimens of Platecarpus planifrons (Cope, 1874) (Squamata: Mosasauridae) and a revised taxonomy of the genus: Journal of Vertebrate Paleontology 27(1): 59-72. 

Lindgren, J., and M. J. Everhart. 2000. Remarks on two problematic mosasaur specimens from the Smoky Hill Chalk (late Cretaceous) of Kansas. Kansas Academy of Science, Transactions Abstracts. 19: 32.

Liggett, G. A., S. C. Bennett, K. Shimada, and J. Huenergarde. 1997. A Late Cretaceous (Cenomanian) fauna in Russell county, KS. Kansas Academy of Science, Transactions Abstracts. 16: 26.

Logan, W. N. 1897. The upper Cretaceous of Kansas; with an introduction by Erasmus Haworth. Kansas Univ. Geol. Survey, v.2:194-234.

Mantell, G. A. 1842. Report on British fossil reptiles. Part 2. Report of the Eleventh Meeting (at Plymouth, 1841), British Association for the Advancement of Science, pp. 60-204.

Martin, J. E., B. A. Schumacher, D. C. Parris, and B. Smith-Grandstaff. 1998. Fossil Vertebrates of the Niobrara Formation in South Dakota. Dakoterra 5: 39-54

Martin, L. D. and J. D. Stewart. 1977. The oldest (Turonian) mosasaurs from Kansas, Journal of Paleontology 51(5): 973-975.

Meek, F.B. and F. V. Hayden. 1861. Descriptions of new lower Silurian (Primordial), Jurassic, Cretaceous, and Tertiary fossils, collected in Nebraska. Philadelphia Acad. Nat. Sci. Proc. 13: 415-447.

Moore, R. C., and W. P. Hays. 1917. Oil and gas resources of Kansas. Kansas Geol. Survey Bull. 3, 391 p.

Nicholls, E. L. 1988. Marine vertebrates of the Pembina Member of the Pierre Shale (Campanian, Upper Cretaceous) of Manitoba and their significance to the biogeography of the Western Interior Seaway. Doctoral dissertation Univ. Calgary, 317 pp.

Novas, F. E.,  M. Fernández, Z. B. Gasparini, J. M. Lirio, H. J. Nuñez and P. Puerta. 2002. Lakumasaurus antarcticus, n. gen. et sp., a new mosasaur (Reptilia, Squamata) from the Upper Cretaceous of Antarctica. Ameghiniana, 39(2):245-249.

Obradovich, J.D. 1993. A Cretaceous time scale. Pages 379-396 in Caldwell, W. G. E. and E. G. Kaufmann, eds., Evolution of the Western Interior Basin, Geol. Assoc. Canada. Spec. Paper 39.

Owen, R. 1851. Monograph on the fossil reptilia of the Cretaceous formations. Palaeontographical Society, London, 118 pp.

Russell, D. A. 1967. Systematics and morphology of American mosasaurs. Peabody Museum of Natural History, Yale University Bulletin 23, 241 pp.

Russell, D. A. 1970. The vertebrate fauna of the Selma Formation of Alabama; Part VII, The Mosasaurs. Fieldiana: Geology Memoirs. 3(7):369-380.

Russell, D. A. 1988. A check list of North American marine Cretaceous vertebrates including fresh water fishes. Tyrrell Mus. Palaeontology, Occasional Paper, 4:23-26.

Russell, D. A. 1993. Vertebrates in the Western Interior Sea. Pages 665-680 in Caldwell, W. G. E. and E. G. Kaufmann, eds., Evolution of the Western Interior Basin, Geological Association of Canada. Special Paper 39.

Sheldon, M. A., 1995. Ontogeny, ecology and evolution of North American Mosasaurids (Clidastes, Platecarpus and Tylosaurus): Evidence from bone microstructure. Unpubl. dissertation. University of Rochester. 183 pp.

Sheldon, M. A. 1996. Stratigraphic distribution of mosasaurs in the Niobrara Formation of Kansas. Paludicola 1: 21-31.

Sheldon, M. A. 1997. Ecological implications of mosasaur bone microstructure. Pages 333-354 in Callaway, J. M. and E. L. Nicholls, eds., Ancient Marine Reptiles. Academic Press, San Diego, CA.

Schumacher, B. A. 1993. Biostratigraphy of Mosasauridae (Squamata, Varanoidea) from the Smoky Hill Chalk Member, Niobrara Chalk (Upper Cretaceous) of Western Kansas. Masters thesis, Fort Hays State University, 68 pp.

Schumacher, B. A. 1997. Lateral stratigraphic and paleontological variation within the Niobrara Formation (Coniacian-early Campanian) of southwestern South Dakota. Doctoral dissertation, South Dakota School of Mines and Technology. 199 pp.

Sternberg, C. H., 1922. Explorations of the Permian of Texas and the chalk of Kansas, 1918. Kansas Academy of Science, Transactions 30(1):119-120. (Papers - Fifty-first annual meeting, 1919), State Printer, Topeka.

Stewart, J. D. 1988. The stratigraphic distribution of Late Cretaceous Protosphyraena in Kansas and Alabama; geology, paleontology and biostratigraphy of western Kansas. Pages 80-94 in Nelson, M. E., ed., Articles in Honor of Myrl V. Walker, Fort Hays Studies, Third Series, No.10, Science.

Stewart, J. D. 1990. Niobrara Formation vertebrate stratigraphy. Pages 19-30 in Bennett, S. C., ed., Niobrara Chalk Excursion Guidebook. Univ. Kansas Mus. Natural History and Kansas Geol. Survey.

Williston, S. W. 1897. The Kansas Niobrara Cretaceous. University Geological Survey of Kansas. 2: 235-246.

Williston, S. W. 1898. Addenda to Part I.   The University Geological Survey of Kansas, 4: 28-32.

Williston, S. W. 1898. Mosasaurs. Univ. Geol. Survey Kansas, Paleontology. 4(5): 81-221.

Zakrzewski, R. J. 1996. Geologic studies in western Kansas in the 19^th Century. Kansas Academy of Science, Transactions 99(3-4): 124-133.