ã 2004 Che Gwin

The Lance Formation in eastern Wyoming contains some 2000-2500 feet of fluvial mudstones and sandstones. These deposits are rich with vertebrate fossils, including many dinosaur fossils. Remains include isolated and worn bones and teeth, skeletal associations, and bonebeds. Although articulated skeletons are uncommon, their preservation is excellent.

Study of these bone deposits can provide valuable information into the early ecological conditions of the fossil site. Through the taphonomic study of the fossils, we can use the present evidence to predict past conditions of the environment in which the bones were buffed. Predictions of fluvial currents, watering holes, migration routes, and carnivorous activities in the past can be deduced. Specimens collected in the summer of 1997 were studied in order to determine whether or not there was a preferred orientation to the bone deposition. The data (angle of deposit, quadrant, direction of deposit) was entered onto a stereographic (stereonet) projection. These projections allow three-dimensional plots to be graphed two dimensionally. They indicate the type of bone orientation present in the site in which they were collected. The results show a stereograph projection with dots mainly in the northwest-northeast quadrants. The long bones plotted are found primarily in the northeast quadrant. The hypothesis is that there is a preferred orientation of how the fossil bed was deposited. The preferred orientation suggests a fluvial current that once ran through the dig site and a rapid deposit of the bones in that dig site.

Fossils are important in reconstructing single organisms. However, they are also important on a larger scale: the reconstruction of the ecosystems of the past. Taphonomic analysis of fossil collections can help in predicting conditions of this past. Taphonomy is essentially "the laws of burial" and concerns all aspects of the passing of biomass from the biosphere to the lithosphere (Olson 1980). Data on bone dimensions and deposition angles can be used in an attempt to determine major geological and biological processes responsible for the site formation and fossil orientation (Fiorillo 1988).

Unless certain conditions are observed, it is unrealistic to assume that where a fossil is deposited is where the organism actually died. Most bones have been transported by forces such as water, wind, scavengers, and carnivores (Shipman 1981). The Hanson Ranch in Roxson, Wyoming, located in eastern-most Wyoming, exhibits extensive outcrops of the Lance formation which contains some 2,000-2,500 feet of fluvial mudstone and sandstone which are good agents for bone preservation.

The Lance formation outcrops up to 20 miles wide along the eastern and western sides of the Powder River Basin (Figure 1). The belt starts near Lance Creek in central Niobrara County and projects through northern Niobrara, Weston, Crook, and Campbell counties. (Love & Christiansen 1985). Hanson Ranch is located in northeastern Niobrara county.

The Lance formation is underlain by the Fox Hills deposit and is overlaid by the Fort Union formation. This dinosaur-bearing strata is representative of the late Cretaceous period. The stratigraphy contains dark, irregularly stratified shales and cross-stratified sandstones in the lowest part of the Lance formation and the upper 300 feet or so of the Lance formation contains more regularly and thinly bedded shales (Clemens 1964). The sandstones are of particular interest because they contain many fossil vertebrates. However, isolated bones, bone fragments, and on occasion articulated skeletons have been found in the shale layers (Clemens 1964). Dinosaur bones are commonly found in fine-grained sediments and are rarely found in the upper Lance formation. Lower layers in the type Lance formation contain many well preserved bones.

By analyzing the bones collected at the Hanson Ranch dig site, it may be possible to reconstruct past conditions of the ecosystem present at that site. The presence of a distinct, preferred orientation of bones in the fossil bed would be suggestive of the involvement of fluvial processes in the burial of the bones. Upon initial review, it appears that there was indeed a high velocity current that deposited these bones quite rapidly.

In June of 1997, an organized dig on the Hanson Ranch was conducted (See Figure 2 for topography and location). Dr. Lee Spencer, director of the Earth History Research Center at Southwestern Adventist University (SWAU), supervised the excavations. A master grid was formed that ran twelve meters from west to east and sixteen meters from north to south. The grid was broken further into smaller grids, each two meters square. Each 2 x 2m. grid was sectioned off. There were four sections - northeast, southeast, northwest, and southwest. Each of these sections were measured as 1 x lm. (Figure 3).

Each excavator was assigned a numbered section in which to work. The sections were numbered 841-846 from west to east and 123-116 from north to south. The excavator was assigned responsibility under the supervisor to properly measure the bones, carefully excavating them and tagging them for further analysis. Measurements on the bone included the degree of plunge and the angle of the trend with respect to north, the distances from the sides of the quadrant, and the length, width, and height of the fossil. Some of these measurements were impossible to measure so they were left out.

Data sheets were collected from all participants by Dr. Spencer. Personal copies of all the sheets were made for my use. The measurements were transferred into an Excel spreadsheet to make data analysis easier. The horizontal plane of the ground served as the reference direction and the degree of plunge and trend were taken from that plane. These two measurements were then transferred onto a stereonet diagram. The degree of plunge corresponds to longitude and the degree of trend corresponds to latitude.

The final step in the analysis involved comparing the stereonet diagram for the Hanson Ranch to the three basic stereonet diagrams. Two stereonet diagrams were plotted. One plot showed all thirty-five bones used and all measurements were regarded as the same. On the other plot, long bones were considered separately from round or square bones. Long bones were defined as being twice as long as they were wide. These bones were plotted in a different color from the other bones on a second stereonet diagram.

A breakdown of which quadrant the bones were located in was entered into a table. This table allows for a specific review of exactly where each bone was located in the master grid. The trend of each bone was converted into a degree based off of a 360 degree compass. For example, a N25E trend converts to 25 degrees and a N50W trend converts to 310 degrees. Using these converted trends, now called plunge direction, and the plunge angle, a regression analysis test and an ANOVA test were run in order to check for correlations between the plunge direction and the plunge angle. Statistical Analysis was supervised by Dr. Watson Chin.

All of the bone dimensions and degrees of orientation with regard to north were taken. These measurements were then recorded in a spreadsheet in order to more efficiently organize the data to be used (Table I). Unfortunately, some bone information was not recorded correctly and therefore was of no use in the analysis. The reason they were of no analytical value was because they lacked proper trend and plunge measurements. These bones are indicated by a "not used" notation within the spreadsheet.

Thirty-five bones were plotted on the stereonet diagram. Table II gives a quadrant breakdown of the location of each bone. Each quadrant was broken down into three sections by degrees. Percentages were calculated from the total number of bones in each quadrant and entered for statistical purposes (Table II.). Bones with 0 degrees trend were taken to be in the north quadrant. Looking at this table provides a clear indication where most of the bones were found with regards to the north.

Most of the bones have an orientation in the north-northwest quadrant. All bones were counted as having the same analytical influence and were plotted on a stereonet projection (Figure 4). According to this projection, most bones are clustered in the northwest and southeast direction. This would indicate a force, probably fluvial, that oriented these bones in this direction. However, this conclusion is a bit deceiving due to the fact that both small and long bones were considered as having the same influence on data analysis. Voorhies (1969) found that long bones usually orient themselves with their long axes parallel to the current flow. If the bones are partly out of the water though, they have a tendency to become aligned perpendicularly to the current and rolled along the bottom of the bed. The bones are most stable in whichever orientation minimizes the surface area in which drag from the current can act upon. It is important to study the orientation properties of these long bones because they can be clear indicators of taphonomic processes. By plotting the long bones in a different color (blue) than the small bones, quite different conclusions can be drawn. I have defined a long bone as one that is at least twice as long as it is wide. This provides a long axis in which fluvial forces can act upon. Spheric or cubic bones provide little information on fluvial current taphonomy due to the fact that small size does not provide a long axis in which the current can align the bone. Figure 5 shows that most of the long bones have a preferred orientation in the north-northeast quadrant. This suggests that because most long bones are located in the north-northeast quadrant, this is the most stable orientation for these bones. Therefore, it now seems that a fluvial current in the northeast direction deposited these bones. Comparing Figure 5 to the five model distributions defined by Toots (1965a), it can be deduced that a unimodal distribution is the closest fit. A unimodal distribution indicates a unidirectional current force that influences the fossil deposition.

There are several factors that indicate that fluvial processes may have been responsible for the deposition of these bones on Hanson Ranch. Earlier studies of the fossils in the Lance formation suggest fluvial deposition. Fossils found in sandstone resembling channel fillings contain many fish and hygrophilous terrestrial vertebrates. The stratigraphy and make-up of these sediments indicate fluvial origins (Clemens 1964). Meschter (1958) studied similar sandstone deposits in the Wasatch formation of the Powder River basin and concluded that they were formed by the action of moving subterranean water. This explanation appears to be equally applicable to the sandstone deposits of the Lance formation. The Lance formation is also very similar to the areas in lower parts of the Mississippi River valley. Most vertebrate fossils are found in the upper half of the Lance formation. It seems that these vertebrate fossils were probably deposited by fresh or slightly brackish water (Clemens 1964) because marine fossils have rarely been found in the Lance formation. Voorhies (1969), in his work at Verdigre Quarry in Knox County, Nebraska, concluded that fluvial sorting might have accounted for at least some of the observed incompleteness of skeletons at the fossil sites. No complete skeletons were found at the Hanson Ranch dig site during this period.

A Regression analysis test and an ANOVA test were run to test for correlations between the plunge angle and the plunge direction (Table III). Whether there is a correlation between these two measurements or not can suggest what type of fluvial condition was present at the time of burial. The F value for the ANOVA test was 2.828048. This number is considerably lower than the F critical value for the degrees of freedom. Because this number is lower, there is no correlation between the plunge angle and plunge direction. A graph of the plunge direction versus the plunge angle shows no correlation also (Figure 6). The indication that no correlation exists suggests that the burial of the bones was quite rapid and that the current allowed for no preferred deposition angle. The angle of the deposit is independent of the direction of the deposit, indicating that the bones may have been rolling or tumbling during their rapid deposit.

The regression and ANOVA test confirm that there is no correlation between the plunge direction and plunge angle of the fossils. This indicates that there was no preferred angle of plunge based on the orientation of the bones. A high velocity current would provide a situation in which the bones were rolled or tumbled along the bottom of the bed in a random orientation. The stereonet projection showing long bone orientation (Figure 5) indicates that there was a preferred orientation of the long bones in a northeasterly manner. The high velocity current that caused random plunge also lined the axes of the long bones towards the northeast. It can be concluded, based on the data and figures in this paper along with earlier studies on the Lance formation, that a high velocity current ran in a northeasterly direction and deposited these bones catastrophically, or in a single occurrence.