Effects of environmental changes on molluscan evolutionary patterns, Gosport Sand (Middle Eocene), Southwest Alabama

Effects of environmental changes on molluscan evolutionary patterns, Gosport Sand (Middle Eocene), Southwest Alabama

Material Information

Effects of environmental changes on molluscan evolutionary patterns, Gosport Sand (Middle Eocene), Southwest Alabama
Harrison, Henry Clifford
Place of Publication:
Tampa, Florida
University of South Florida
Publication Date:
Physical Description:
vii, 93 leaves : ill. (some col.) ; 29 cm.


General Note:
Thesis (M.S.)--University of South Florida, 1994. Includes bibliographical references (leaves 69-74).

Record Information

Source Institution:
University of South Florida
Holding Location:
Universtity of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
020129368 ( ALEPH )
32603828 ( OCLC )
F51-00111 ( USFLDC DOI )
f51.111 ( USFLDC Handle )

Postcard Information



This item is only available as the following downloads:

Full Text


EFFECTS OF ENVIRONMENTAL CHANGES ON MOLLUSCAN EVOLUTIONARY PATIERNS, GOSPORT SAND (MIDDLE EOCENE), SOUTHWEST ALABAMA by HENRY CLIFFORD HARRISON IV A thesis submitte d in partial fulfillment of the requirements for the degree of Master of Science Department of Geology University of South Florida August 1994 Major Profe ssor: Lisa L. Robbins, Ph.D.


Graduate School University of South Florida Tampa, Florida CERTIFICATE OF APPROVAL Master's Thesis This is to certify that the Master's Thesis of HENRY CLIFFORD HARRISON IV with a major in Geology has been approved by the Examining Committee on May 19, 1994 as satisfactory for the thesis requirement for the Master of Science degree Examining Committee: --Member: Richard A. )ia\lis, Jr., Ph.D. /\ Mehlber: Peter J. Harries, Ph.D.


Henry Clifford Harrison IV 1994 All Rights Res erved


DEDICATION To my wife, Holly Reed-Harrison who bas sacrificed more than anyone ever should during my graduate career, and somehow bas bung on through this who l e ordeal. I will forever be indebted to her for putting up with me an d this thesis. To my mother Dr. Elizabeth A. Harrison, who helped me start my first bug collection.


ACKNOWLEDGEMENTS It is with the deepest gratitude that I wish to thank the many people without whom this thesis could never have been completed. Blame for getting me interested in molluscan paleontology lies wholly with Dr. Warren D. Allmon, whose ability to inspire interest in the biology of old, dead snails is without comparison. Nick Tew and Andy Rindsberg of the Alabama Geological Survey provided invaluable understanding of the extraordinarily complete geologic section in the coastal plain of Alabama Holly Reed-Harrison and Eric Reed acted as faithful field slaves during the collecting phase of this study. The drafting skills of Chris Martinez were invaluable in preparing many of the figures, while Brian Peck wrote the computer code that generated the stratigraphic range chart. Financial, material, and technical support for this thesis was provided by the Geological Society of America, the Petroleum Research Foundation of the American Chemical Society the Geology Department at The University of South Florida, and by ViroGroup Inc./Missimer Division. Finally, the greatest thanks must be extended to my friends and colleagues who have assisted me in the most important way by always being there when I needed them and for believing in me Most of all, thank you Holl y


TABLE OF CONTENTS LIST OF FIGURES ABSTRACT 1. INTRODUCTION Statement of Problem Investigative Procedures 2. STUDY AREA Regional Stratigraphic Relationships Facies Relationships Study Sites Stratigraphy 3. METHODS Sample Collection Data Collection 4. RESULTS Glauconitic Grain Maturity Sediment Characteristics Faunal Diversity and Species Occurrence 5. DISCUSSION Background Work Implications of Sediment Characteristics Interpretations of Glauconitic Maturity Sedimentation Rates Faunal Turnover and Diversity Trends lll v 1 1 3 6 6 11 13 15 18 18 20 31 31 35 45 49 49 52 55 61 63




Figure 1. Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. Figure 8. Figure 9. Figure 10. Figure 11. Figure 12. Figure 13. LIST OF FIGURES Map of Study Area and Outcrop Localities in Southwest Alabama Generalized Regional Stratigraphy, Showing the R elationship of the Gosport Sand to the Moody's Branch, Cockfield, and Lisbon Formations Possible Paleogeography of Highly Diverse and Abundant Molluscan Shell Deposits Nutrient Rich Sediment Starved, Open Shelf Areas Between Del taic Depocenter s (After Allmon 1989) Generalized Stratigraphic Section of the Gosport Sand Formation Sample Retrieval Procedure Flowchart of Laboratory Procedures Raref action Graph of a Representative Sample Rank/ Abundance Graph for a Representative Sample Glauconitic Grain Maturity; Little Stave Creek Glauconitic Grain Maturity; Cla iborne Bluff Mud, Sand and Gravel Percentages in R aw Samples; Little Stave Creek Mud, Sand, and Gravel Percentages in Raw Samples; Claiborne Bluff Mud and Sand Percentages of Non-carbonate Sediments; Little Stave Creek 111 7 8 14 16 19 21 25 27 33 34 36 37 38


Figure 14. Mud and Sand Percentages of Non-carbonate Sediments; Claiborne Bluff 39 Figure 15. Mean Grain Size and Standard Deviation of Noncarbonate Sediments; Little Stave Creek 40 Figure 16. Mean Grain Size and Standard Deviation of Noncarbonate Sediments; Claiborne Bluff 41 Figure 17. Sediment Carbonate Content and Mud, Sand, and Gravel Percentages of Carbonate Sediments; Little Stave Creek 43 Figure 18. Sediment Carbonate Content and Mud, Sand, and Gravel Percentages of Carbonate Sediments; Claiborne Bluff 44 Figure 19. Vertical Distributions and Abundances of Species that are Common in at Least One Sample 47 Figure 20. Sample Sizes, Number of Common Individuals per Sample, Number of Common Species per Sample, Species Richness, and Evenness; Claiborne Bluff 48 Figure 21. Photomicrographs (Crossed Polars) of Mature and Immature Glauconitic Grains 60 Figure 22. Generalized Relative Sea Level Change During the Middle to Late Eocene in the Eastern Gulf Coastal Plain (After Mancini and Tew, 1991) 62 IV


EFFECTS OF ENVIRONMENTAL CHANGES ON MOLLUSCAN EVOLUTIONARY PATTERNS, GOSPORT SAND (MIDDLE EOCENE), SOUTHWEST ALABAMA by HENRY CLIFFORD HARRISON IV An Abstract Of a thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Geology University of South Florida August 1994 Major Professor: Lisa L. Robbins, Ph.D. v


Highly diverse molluscan fossil assemblages are a common occurrence in the Gulf Coastal Plain in sandy, glauconitic deposits The origin of t he high diversity in these shell beds is enigmatic; high diversity in fossil accumulations can be derived from stable, long-lasting, highly diverse communitie s or from rapidly changing community st ructures. The hypothesis that the cause of the high diver sity (495 species) in the Middle Eocene Gosport Sand Formation, southwestern Alabama was due to the time-averaging of successive, differing communities was examined through three independent lines of evidence: sediment grain-size distribution, trends in glauconitic grain maturity and changes in two related indices of diver sity up-section. Non-carbonate grain size remained constant upse ction while carbonate (shell) fragments bec a me significantly smaller toward the top of the formation. In the sa me fashion, glauconitic grains demonstrate maturation up-section. These patterns suggest a slowing in sed iment a tion rates as the Gosport Sand was dep osited. Such a redu ct ion in se diment accumulation would "condense" the fossil record resulting in a fossil assemblage that was more diverse than the liv ing community from which it was derived. Statistical mea sures of species richness and evenness did not show significant changes up-section, sugges ting stabi lity in the community structure, but the actual patterns of occurrence of the common species were so highl y variable as to s ugge st th at living communities were in a constant state of flux. The successive assemblages grade into one another, with species becoming more and less dominant at s poradic intervals. VI


It is apparent from these patterns that, while evidence of rapid and drastic environmental changes was not recorded in the Gosport Sand the time-averaging of the fossil remains of the succession of rapidly changing community structures resulted in a fossil deposit with a greater apparent diversity than the living community ever displayed at any one time. Temporal dynamics of the structure of the living communities, coupled with a slow rate of sediment accumulation, were the source of the high levels of diversity observed in the Gosport Sand. Abstract Approved : Major Professor: Lisa L. Robbins, Ph.D. Associate Professor Department of Geology Date Approved: VII


1 1. INTRODUCTION Statement of Problem Shelly, glauconitic sands are common in the Lower Tertiary deposits of the Gulf Coastal Plain. Preserved within some of these deposits are molluscan fossil assemblages that often contain many times the number of species in a given formation than are generally found in modern communities (Gardner, 1957). One s uch fossil deposit is present within the Middle Eocene Gosport Sand Formation in southwestern Alabam a The Gosport Sand ranks as one of the most diver se fos sil accumulations in the entire Gulf Coastal Plain as it contains as many as 495 s pecies of molluscan fossils (Palmer, 1937 ; Palmer and Brann, 1965, 1966; Allmon 1988). A random sample at any g iven interval in the Gosport Sand may contain upwards of 100 molluscan species (Cobabe and Allmon, 1993). The reason for the high diversity observed in the Gosport Sand is enigmatic. Immigration and emigration of s pecies from an area can result in highly diverse fossil accumulations. Patterns of high diversity throughout a given formation may also be a result of changes in environment and associated faunal adaptations, which could cause rapid rates of community structure overturn, and therefore hi g h incidences of speciation and extinction (Simpson, 1944).


2 Alternatively, Slobodkin and Sander's (1969) time-stability hypothesis suggests that diversity levels should be greatest in the most stable and longest-persisting communities. Two hypotheses are proposed to explain the high diversity observed in the Gosport Sand: 1) the high mollu scan diversity in the Gosport Sand may be due to community overturning in a constantly shifting environment with time averaging of successive communities resulting in diversity levels in the fossil assemblage that are higher than those that actually existed in the original communities, or 2) the Gosport Sand was accumulated as a result of a series of very diverse, long-ranging and stable communities. A determination of the origins of the high diversity in the molluscan fossil assemblage preserved in the Gosport Sand will allow a better understanding of the reasons for the wide range in diversities in fossil assemblages across the Gulf Coastal Plain. Allmon (1988, 1989) showed that densely fossiliferous shell beds, while common in the Gulf Coasta l Plain, occur predominately in glauconitic sands and in limestones that display considerable terrigineous influence Non glauconitic sands and marine limestones generally contain far fewer species often by orders of magnitude (Dockery, 1986). This pattern may be an artifact of differing degrees of preservation or may be related to the true diversity in the original molluscan communities. The Gosport Sand, as one of the most diverse accumulations of molluscan fossils in the Gulf Coastal Plain (Toulmin, 1977), offers an excellent opportunity to examine the factors that contributed to the accumulation of a highly diverse


3 .fossil shell assemblage. The complete exposures of the Gosport Sand, accessible along rivers and creeks in southwestern Alabama allow comparison of sedimentologic geochemical, and paleontologic trends in a single continuous section, providing new insight into the reasons for the high diversity in many other Tertiary deposits. Investigative Procedures Changes in species diversity independent of which particular species are actually present, can be diagnostic of changes in the environment of deposition of recent and ancient shell deposits (Kornicker and Odum, 1967). If environmental fluctuations resulted in changes in community structure, the time averaging of the successive communities could result in high diversity such as that observed in the Gosport Sand. It would be expected that changes in diversity and composition of the fossil assemblages in successive samples would coincide with variations in sedimentological indicators of the environmental fluctuations. To test this hypothesis, samples were collected in a continuous fashion from the base to the top of the Gosport Sand at two locations in southwestern Alabama. Statistics for species abundance (diversity) and evenness (the reciprocal of dominance) were calculated for each sample, and the fossil assemblage in each sample was identified Stratigraphic occurrences of the species were tabulated along with their abundances in each sample. From this database, the occurrences of the 43


4 species that are common in the Gosport Sand were extracted and plotted to compare the observed evolutionary patterns with two easily measured sediment characteristics, grain size and sorting. Changes in these sedimentary measures can reflect changes in the wave and current energy of the depositional environment (Gross, 1982). Changes in grain size and sorting therefore can be considered suitable proxies for environmental changes (Tanner, 1982) Another, less commonly utilized environmental indicator is the extent to which the glauconitic component of the sediment has fully developed (Hower, 1961). A common constituent of sediments in the Gosport Sand glauconitic grains change in appearance as they evolve during their residence time in a s hallow zone below the sediment-seawater interface (Odin and Matter 1981 ; Mallinson 1988). If the rate of burial slows either due to reduced sediment load, increased bioturbation, or waveand current-related sediment reworking, the gr a ins remain in this zone of formation for longer periods of time, becoming more and more mature. Glauconitic grains are easily classified as either mature or immature, depending on their physical appearance. As grains mature they evolve from a porous light green state to a denser, rounded, dark green to black grain (Ireland et al., 1983). This process is gradational throughout the glauconitization process and can be halted at any time by removal from the zone of glauconitization. It is not clear whether the maturation process can be re-activated if the grains are returned through reworking into a zone suitable for glauconite formation but


. sediment reworking that introduces the glauconitic grains into an oxidizing environment will result in weathered, oxidized grains (Odin and Matter, 1981). Changes in the degree of maturation of glauconitic grains in the Gosport Sand were used to examine environmental stability (as defined by changing sedimentation rates and relative degrees of sediment re-working). If rates of sedimentation and burial had remained fairly constant throughout the deposition of the Gosport Sand, glauconitic maturity should remain relatively constant. The degree of glauconitic maturity in successive samples was compared to the fossil faunal assemblage patterns and the sediment grain-size measurements, providing the basis for interpreting faunal changes with respect to changes in the depositional environment. The relationships between the three classes of data were then analyzed to determine the source of the high molluscan diversity reported from the Gosport Sand. 5


6 2 STUDY AREA Regional Stratigraphic Relationships The Gosport Sand Formation is a late Middle Eocene (end Claibornian) deposit of shelly, glauconitic, fineto medium-grained sands, with scattered thinly laminated carbonaceous clays at the top of the section at at least one locality. It crops out along an east-west trend for approximately 90 km along strike across southern Alabama (Figure 1) and ranges from 3 to 12 meters thick. Exposures of the Gosport Sand, including the type locality at Gosport Landing on the Alabama River are concentrated in the southwestern portion of the state. Several p a rtial exposures are reported in early literature but extensive field reconnaissance failed to locate any significantly complete exposures other than as noted on Figure 1. Across the outcrop area, the Gosport Sand Formation unconformably overlies the Middle Eocene Lisbon Formation (Figure 2). This contact is highly burrowed, with Ophiomorpha extending as deep as 50 em into the underlying Lisbon Formation. The unconformity between the two has been estimated to represent a period of 0 6 million years, based on graphic correlation of micro -and macrofossil occurrences, while the duration of the deposition of the Gosport Sand is estimated at 1.5 million years (Hazel et al., 1984). The Gosport Sand is


0 KM B I R!.!INGHAM 100 !.IONTGOM ER Y \ JACKSO N Figure 1. M a p of Study Area and Out Crop Locations in Southwest A labama. 7 1 Puss C uss Creek; 2 Silas; 3 Gopher Hill; 4 LilLie Stave C reek; 5 -Gosport Landing 6 C l a iborn e Bluff; 7 -Rattlesnak e Bluff.


MISS ISSIPPI __ w M O O D Y S BRANC H fOR MATIO N 8 G EOR G I A -E lr---3 fORMATION GOSPORT SAND h ----56_ .... ( l 2 0 Figure 2. KM 100 LISBON fOR M ArtON General i zed R eg iona l Stratigraphy, Sho w ing the R e lationship of the Go sport Sand to the Moody's Branch, Cockfield, and Li s bon Formations. Numbers refer to the a pprox i mate location and interval exposed at each lo ca lity (see Figure 1 ).


overlain by the Jacksonian-age Moody's Branch Formation. Siesser et al. (1985) assigned this boundary to the calcareous nannoplankton zone NP17 approximately 45 Mya 9 Near the Alabama-Mississippi state line the Gosport Sand interfingers with the lignitic clays of the Middle Eocene Cockfield Formation (Blanpied and Hazzard, 1938; Mancini and Tew, 1989). This facies relationship has been recorded in cores and at several outcrops (Blanpied and Hazzard, 1938). The Gosport Sand is not recognized in Mississippi; the equivalent in that state is considered to be part of the 15 to 140 meters of cross-bedded deltaic sands and carbonaceous clays of the Cockfield Formation (Toulmin et al., 1951; Mancini and Tew, 1989). To the east of the Alabama River, the Gosport Sand is not recognizable in outcrop, althoug h glauconitic sands along the Conecuh River near Andalusia and along the Chattahoochee River have been assigned to the formation by at least one author (MacNeil, 1946). Toulmin and LaMoreaux (1963) rejected the eastward extension of the Gosport Sand to the Alabama-Georgia border, regarding the Gosport Sand as pinching out between Little Stave Creek and the Chattahoochee River. The similarity of the fossil molluscan faunas with those of the Moody's Branch Formation in southwestern Alabama caused them to regard the sediments directly overlying the Li s bon Formation along the Chattahoochee as be l onging to the Moody's Branch Formation. Lithologic similarities between these and the Moody's Branch Formation exposures to the west strengthened their case; in both


10 areas the Moody's Branch Formation is typified by shelly, sandy limestone, compared to the shelly sands of the Gosport Sand. The hypothesis that the Gosport Sand pinches out in central to eastern Alabama could not be reinvestigated due to poor fossil preservation and the drowning of exposures by reservoir construction. Recent work by Hazel (pers. cornm.) on ostracode faunas however, indicates that the lower Moody's Branch Formation on the Conecuh River may at least be time-equivalent to the Gosport Sand. A Claibornian age for the Gosport Sand has been a source of much disagreement throughout the latter half of this century. The Claibornian Jacksonian boundary has been placed at different times and by different authors between the underlying Lisbon Formation and the Gosport Sand (Hardenbol and Berggren 1978) or between the Gosport Sand and the overlying Moody s Branch Formation (Bybell, 1975; Siesser, 1983; Dockery, 1984, 1987) A Claibornian age has been retained for the Gosport Sand based upon similarities between the molluscan and microfossil assemblages in the Gosport Sand and underlying formations (Berggren et al., 1985); only 45 species from the Gosport Sand occur in the overlying Jacksonian Moody's Branch Formation but 145 species found in the Gosport Sand originated prior to its deposition (Dockery 1984). On a more regional scale, the McBean Formation in Georgia has been correlated with the Gosport Sand (Huddleston, 1965). In Florida, the Avon Park Limestone has been correlated with the uppermost part of the Lisbon Formation and the Gosport Sand (Toulmin 1977). The Moody s Branch Formation is


11 considered to be equivalent to the Ocala Limestone which also outcrops in Florida (Vernon, 1951). To the west of Mississippi the time-equivalent section expands into the thick deltaic sequences of the Yegua, Cockfield, and Cook Mountain formations (Blanpied and Hazzard 1938 ; Davies and Etheridge 1971; Galloway, 1989). Facies Relationships Hazel (pers comm.), using ostracode distributions, and Bandy (1949), using foraminifera, detected no gap in either taxon's evolutionary lineage across the Gosport Sand-Moody's Branch Formation boundary at Little Sta ve Creek, suggesting nearly constant deposition across the boundary. In contrast, Bornbause r (1947) and later Rainwater (1968) observed that the late Middle Eocene Cockfield progradational deltaic deposits which interfinger with t he Gosport Sand are unconformably overlain by the basal trans g res sive deposits of the early Upper Eocene Moody's Branch Formation. MacNeil (1946) identifi ed both the Lisbon-Gosport Sand and the Gosport Sand-Moody's Branch contacts as transgressive disconformities on the basis of several features : 1) uneven erosional surfaces, 2) basal conglomerates mostly of local origin, 3) basal glauconitic sand or marl and 4) burrows in basal glauconites extending downward i nto underlying units. Mancini and Tew (1990) concurred by placing a Type 1 unconformity at the Gosport-Moody's contact on the basis of quartz pebble l ag and the change in


12 lithologies. This study failed to verify Mancini and Tew's lag deposit at the study sites, but a sharp, burrowed, lithologic transition from shelly sand (Gosport Sand) to limestone (Moody's Branch Formation) occurred at each outcrop except Little Stave Creek. The base of the Moody's Branch Formation at Little Stave Creek is a shelly, sandy, weathered limestone. The apparent discrepancy between Hazel and Bandy's conclusions (that the Moody's Branch conformably overlies the Gosport Sand) and those of other investigators (that an unconformity exists between the two) may be facies related. Little Stave Creek, where Hazel and Bandy focused their studies, is near the center of the Gosport outcrop belt and may have experienced uninterrupted deposition during times in which the eastern and western margins may have been subject to nearshore erosion Evidence of nearshore deposition (thinly laminated lignitic clays) exists in at least one exposure of the sediments just below the contact between the Gosport Sand and the Moody's Branch Formation at Claiborne Bluff; these clays are absent at the Gosport Sand Moody's Branch Formation gradational contact at Little Stave Creek. Any paleodepth differences across the outcrop area must have been slight, however, because fossil communities are similar at all exposures Davies and Etheridge (1971) characterize the Middle Eocene of the Gulf Coastal Plain as being a time during which sediment sources were frequently shifted laterally along the coast due to the migrations of large fluvial, deltaic, and interdeltaic complexes on the margins of a gradually subsiding basin. The changes


13 in the loci of the sediment supply would have been accompanied by changing sedimentologic environments shelfward of the deltaic complexes. The Gosport Sand was deposited on the eastern flank of the Mississippi Embayment (Mancini and Tew, 1988), in a region that would have been influenced by runoff from the Appalachians (Hansen, 1987) This proximity to both open marine and deltaic/ estuarine environments fits the paleogeographic model proposed by Allmon (1989) for the densely fossiliferous glauconitic shell deposits of the Gulf Coastal Plain: nutrient-rich, sand-starved, open shelf areas between deltaic depocenters (see Figure 3). Study Sites Outcrops were examined at seven sites (Figure 1). From west to east, they are: 1) a road cut 0.6 km east of Puss Cuss Creek on County Road 14, between Melvin and Gilbertown Choctaw County, 2) a road cut on the north side of U.S. Highway 84, 6.9 km west of Silas, Choctaw County, 3) Gopher or Baker's Hill, 1.2 km above St. Stephen's quarry on the Tombigbee River, Washington County, 4) along Little Stave Creek, a westward flowing tributary of Stave Creek 5.6 km north of Jackson and west of U.S. Highway 43, Clarke County 5) Gosport Landing (the type locality of the Gosport Sand Formation), at a high bluff on the right bank of the Alabama River, 7.2 km below the bridge on U .S. Highway 84, 6) Claiborne Bluff, at a high bluff on the left bank of the Alabama River 0.4 km


F i g ur e 3. \ : / It\..."{ --------------OELT A 1>-cc"O'f..)\\J.QOI>-v\J. H ICH SE D IMEN TATION r r" H IGH NUTR I ENTS ,ooDELTA H IGH SEOIM ENTA TION HIGH NUTRIENTS LOW SEDIMENTATION HIGH NUTRIENTS PROPOSED AREA OF GOSPORT-TYPE SHELL BED FORMATION GULF OF J v!EX ICO 14 Possi ble P aleogeograph y of Hi ghly Dive rse a nd Abundant M o llu scan Shell Deposi t s Nu t rien t Ric h Sediment Starved Open Shelf Areas Between Delt a i c Depocenters (After Allmon, 19 89).


15 below the bridge on U.S. Highway 84, and 7) Rattlesnake Bluff on the left bank of the Alabama River 10.5 km below the bridge on U.S. Highway 84. A composite section of the Gosport Sand is depicted in Figure 4. Lithologic descriptions of each of the seven outcrops are presented in Appendix 1. The relative position of each of the exposures within the regional stratigraphy is depicted in Figure 2. Localities # 1 and #2 are only partial exposures, exposing only the top one to two meters of highly weathered Gosport Sand sediments and were therefore not sampled for this study. Outcrop localities #3 and #7 are complete exposures, but the base of the Gosport is just above water level at low river stages and the outcrops are flooded severa l times a year. This results in poor fossi l quality due to dissolution by ri ver water. Outcrop #5 was inaccessible excep t at the base of the Gosport Sand. Samples for this study were retrieved from outcrops #4 and #6 (see "Methods" for sample collection techniques). The relationships between each of the exposures and the formation s studied are depicted in Figure 2. Stratigraphy The only se dim entary structures observed in outcrop were Ophiomorpha that are ubiquitous at the base of the Gosport and common at mid-section. The low angle planar cross-bedding that was noted at Gosport Landing by Swindel ( 1 986) was not observed at any of the exposures visited. The only stratification s




observed were thick (50 80 em) bands of slightly varying colors, apparently related to river level fluctuations 17 The indurated calcareous sandstone ledge that occurs midway up-section in outcrops #3, #4, and #5 is possibly the result of diagenetic processes The Ophiomorpha occur within three feet above and below this indurated ledge, and the faunal assemblage within this ledge is generally s imilar to the assemblages in the layers above and below it. The non-carbonate sediments in the sandstone are similar in grain-size distribution to overlying and underlying strata, as well. At Little Stave Creek the ledge contains a concentration of the oyster Crassostrea alabamiensis (Lea) but no other exposure of the ledge contains this taxon The almost total absence of any other identifiable mollusks and the high degree of dissolution of the heavily bioeroded, well cemented oyster shells suggests that the indurated ledge may be the result of the action of meteoric waters. There is no fauna that is unique to this ledge at more than one exposure, suggesting that the control on the cementation of this ledge was a regional diagenetic phenomenon possibly due to the location of the water table at some point in time. Thin (2-3 em) stringers of limestone and clayey sands at the top of the Gopher Hill section may be the result of the diagenetic action of groundwater ; geochemical analysis of the cements would be necessary to verify this They are repetitive for almost 1 meter interbedded with 5 10 em of Gosport-type shelly sands and are lithologically similar to the overlying biomicrites of the Moody's Br a nch Formation (see Figure 2).


18 3. METHODS Sample Collection Contiguous samples were collected from two outcrops (Little Stave Creek and Claiborne Bluff) in a continuous series up-section. Rappelling gear was used to descend the bluffs. The outcrop face was measured and described a 10 em wide rind of weathered sediments was removed from a 50 em wide strip from top to bottom of the formation, and 30 em increments were marked on the face. Using a hoe-pick, a vertical groove was carved on each side of the cleared strip, about 10 em deep. A horizontal groove of similar depth was then carved at each 30-cm mark creating a ladder-like pattern (Figure 5) Finally the blocks of sediment between the "rungs" of the ladder were quarried out of the outcrop face as intact as possible, resulting in samples of between 2000 and 10,500 cm3 The variation in sample sizes is due to the fact that the samples often disaggregated as they were being put into the sample bags Sampling proceeded from the base of the Gosport Sand toward the top of the formation in order to minimize mixing of faunas from overlying layers with those of the lower samples.


Figur e 5 1 9 BRANCH FOR MATION I I I I jl I I 30 I C M y BLOCKS Q UARRIED I I GOSPOR T OUT BETWEE N I I SAND GROOVES I 1 30 CM I I I I LISBO N FORMATION Sample Retrieva l Procedure. B l ocks of se diment we r e quarried out of the exposure face w it h a hoe pick t hen placed i nto large samp l e bags as intac t as pos s ib l e.


20 Data Collection The samples were subdivided in the lab for processing (see Figure 6). For grain-size and glauconitic grain maturity analyses, a 1000 cm3 sub-sample was removed in a random grab. 300 glauconitic grains were picked from each raw sediment sample. These grains were classified as immature or mature, based on Odin and Matter's (1981) description of the maturational stages of glauconitization. The abundances of mature grain types in each sample were then plotted in order to determine if there are stratigraphic changes in the maturity of the glauconitic grains in the Gosport sediments. After the glauconitic grain count was completed, the sub-sample was processed through a random sample splitter down to 50 100 grams for grain-size analysis. The raw sample was first weighed and then sieved for three minutes in a Ro-Tap machine through standard Wentworth sieves (-2.0-4.0 phi, 0.5 phi increments). Three minutes is less than the standard sieve time of ten minutes duration, but periods longer than three minutes resulted in significantly increased shell breakage. The shortened sieve time therefore resulted in an underestimation of the pan (mud) fraction of the sediments. Each fraction was then weighed to determine the size-frequency distribution of the sediment sample The raw sample was then re-combined in a beaker, and the calcium carbonate components were completely digested with 10% HCI. After thorough




22 washing and careful decanting, the sample was dried at 50 C and then re-weighed to determine the percentage of calcium carbonate in the raw sample The non carbonate fraction was then resieved to determine its size-frequency distribution These figures were subtracted from the values for the raw sediment to arrive at the approximate size-frequency distribution of the carbonate fraction that was dissolved. Although some carbonate cement was dissolved which had held together finer, non-carbonate grains microscopic examination revealed that cemented grains comprised a small ( < 1%) component of the raw sample. The data collected were utilized to construct size-frequency diagrams and to determine the arithmetic mean and standard deviation of each sample following the methods of Krumbein and Pettijohn (1938). The individual size-frequency diagrams and grain size statistics were then examined for trends through the section. For the faunal analysis, only samples from Claiborne Bluff were examined The work effort required to adequately sort and identify all specimens in the samples was so great that time constraints precluded the thorough processing of the samples from Little Stave Creek. All samples collected for this study are now available for examination at the Paleontological Research Institute in Ithaca, NY. Each bulk sample was dried at room temperature and allowed to disaggregate naturally in order to minimize post-collection breakage of the fossils. Due to the extraordinary abundance of fossil types that have been studied from the Gosport ranging from microscopic calcareous nannoplankton (Bybell 1975) and foraminifera (Bandy, 1949; Golden 1969) to macroscopic mollusks (Palmer, 1937;


23 Palmer and Brann, 1965 1966; and Toulmin 1977), it was necessary to limit the size fraction that would be examined in the faunal study The bulk samples therefore, were sieved through a 3 mm screen. Visual examination of the < 3 mm fraction revealed that an estimated 1 -5 % of the total number of mollusks ( < 1 % of the species) in the bulk samples were lost through the screen The majority ( > 95% ) of the < 3 mm fraction was quartz sand, glauconitic grains, shell fragments, and microfossils (foraminifera ostracodes calcareous nannoplankton, and micromollusks ). The coarse ( > 3 mm) fraction was thoroughly examined and all individual molluscan fossils in the sample were removed for identification For the purposes of this investigation an individual was considered to be: 1) a specimen with > 50 % preserved (bivalves and gastropods); 2) a fragment with length .2:. 10 x diameter (scaphopods) The total number of bivalve shells retrieved was divided by 2 to obtain the number of individual bivalves in the s amples. All fossils picked were identified to species level, if possible, or at least to the genus level, according to Toulmin (1977) and Palmer (1937). The taxonomic nomenclature used is from Palmer and Brann (1965, 1966) and Allmon (1988) In order to examine changes that may have taken place in the community structure through the period in which the Gosport Sand was deposited, the abundances of all molluscan species were tabulated for each sample (see Appendix 2). Despite the large number of species recovered, the majority of the species were represented by one or a few individuals (rare species) Common


24 species (as defined below) made up the minority of the species recovered. Cobabe and Allmon (1993) concurred with Koch (1987) when they concluded that observed stratigraphic ranges can only be considered reliable for common species that are adequately sampled. In order to ascertain whether or not the samples were sufficiently large to yield reliable information, a frequency distribution, or rarefaction, graph depicting the numbers of species with increasing numbers of individuals was constructed for each sample (Figure 7). Numbers of individuals recovered are plotted on the x-axis, and the numbers of species are represented on the y-axis. This depicts the rate at which new species are found with an increasing sampling effort. Similar patterns for each sample suggest that the samples were of sufficient size to overcome possible significant bias due to sample size differences (Magurran, 1988). The similarity of these patterns for each sample also verifies that each sample retrieved sampled the assemblage to approximately the same degree of completeness. Buzas et al. (1982) and Cobabe and Allmon (1993) have shown that in addition to the requirement of sampling completeness, information regarding the stratigraphic ranges of species cannot be considered reliable for species that are not common. They considered a species to be common in a given formation if the species occurred in more than 80% of the samples retrieved from that formation. The applicability of their guidelines to an intraformational study such a s this is less than certain however. In studies such as those conducted by Bu zas et al. (1982) and Cobabe and Allmon (1993), which examined species ranges


0 w .._ z w (f) w a: 0... w a: (f) w 0 w 0... (f) Figure 7. 25 #SPECIES/INDIVIDUAL SAMPLED SAMPLE CL-10 50 40 ,90% 30 80% 20 10 0 1 00 200 300 400 500 600 700 800 900 1 000 # INDIVIDUALS RECOVERED Rarefaction Graph of a Representative Sample. Note that the first 30 % of the sampling effort produces 80% of all species recovered from the sample. Rarefaction curves for all samples displayed similar patterns, suggesting that all samples were sufficiently large to overcome any bias due to sample size inconsistencies.


26 across formational boundaries, conservative estimations of species ranges such as would result from excluding rare species are desirable, since some degree of reworking of fossils from a lower unit is usually encountered at the base of an overlying formation. In a study such as this one, which examines the variations in abundance of species with potentially rapid faunal overturn throughout a continuously deposited formation, excluding species which occur in fewer than 80% of the samples would automatically disregard all species that do not range through at least 80% of the formation. This would mean disregarding all but six of the species recovered from the Gosport Sand. In order to more completely depict the stratigraphic ranges of species, a broader definition of "common" was required. Species were defined as common enough to be included in the stratigraphic range chart by ranking them in order of their abundance in each sample (number of individuals of a species in a sample). A graph was prepared for each sample depicting the number of individuals found in each species (Figure 8). A cut-off point on the curve was chosen that coincides with the point at which the number of individuals was at least 10% of the maximum number of individuals in any species. That is to say, species that are considered "common" were represented by a number of individuals greater than one-tenth of the maximum number for any species in the sample. The stratigraphic distributions and abundances of the 43 species that were common in at least one sample were then plotted to portray faunal turnover through time.


1 80 160 140 (f) 120 _J <( 100 0 > 0 80 z 60 40 20 0 0 Figure 8. 27 #INDIVIDUALS/SPECIES SAMPLE CL1 0 ---------------------------, COMMON SPECIES N(max ) / 1 0 RARE SPECIES 5 1 0 1 5 20 25 30 35 40 4 5 50 55 60 SPECIES RANKING Rank/ Abu n dance Graph for a Representati ve Samp l e Common species are defi n e d as those t o the left of the break point of the curve; for every sample this point i s 1 0 % of the maximum number of individuals in a n y species in the samp l e


28 Studying changes in the composition of the fossil assemblage is one way to examine evolutionary patterns that may have been controlled by environmental conditions; changes in species richness and evenness (or its reciprocal, dominance) may also indicate changing environmental conditions (Kornicker and Odum, 1967; May, 1975) To this end, the Margalef and Berger-Parker indices of diversity (Magurran, 1988) were calculated for the common species, giving a measure of the species abundance and dominance components of diversity. The Margalef index (DMg) is calculated as follows: DMg = (S-1)/ln N 1) in which S = the number of common species recorded in a particular sample and ln N = the natural log of the total number of individuals of common species in the sample (Magurran, 1988). This index has the advantages of giving a quick measure of species richness (number of species per sample), while retaining the ability to discriminate between samples of different sizes and having differing numbers of species A shortcoming of this index is that it is not indicative of the degree of dominance displayed in the frequency distribution graph. Two hypothetical samples, each with 50 species and 100 individuals would have the same value for DMg even though one sample might have 2 individuals in each species, while the other sample might have 1 individual in each of 49 species and 51 individuals in the 50th species.


29 While this hypothetical situation is unlikely to arise in practice, it is advisable to choose another index to complement one that only measures species richness (Magurran, 1988). The Berger-Parker index (d) is an easily calculated measure of the proportional importance of the most abundant species (dominance). This index is calculated by the equation: 2) in which Nmax = the number of individuals in the most abundant species, and N = the total number of individuals in the sample (Magurran, 1988). The reciprocal of this dominance index is often used since an increase in diversity, or evenness, is indicated by an increase in 1/d. The Berger-Parker index is influenced by samp le s i ze, but is independent of species richness and so is considered to be one of the most sat isfactory dominance measures available (May, 1975) The sensitivity to sa mple size is minimi ze d when a majority of species present in a single sa mple are recovered and when individuals in a species are not grouped into aggregations (Magurran, 1988). Although the samples collected for this study were not equal in size, meaningful results can still be derived from this fossil assemblage. No aggregations of individuals of one species were observed, satisfying one of the cautions on the use of the Berger Parker index for samples of unequal s i ze. A majority of the species present may not have been recovered, as Cobabe and


30 Allmon (1993) have shown that it would be necessary to coll e ct samples of 25, 000 to 55, 000 cm3 in order to adequately sample moderately common and rare specie s in the Gosport Sand. Samples such as the ones used in this study ( 2000 to 10 500 cm3), therefore, can only supply adequate abundance and stra t igraphic information for the most abundant species, supporting this use of the Berger-Parker index i n the comparison of diversity in successive samples. The final step in the analysis of diversity trends in the Gosport Sand was to plot the indices of species richness (DMg) and evenness (1/ d) for each sample up the s ection These plots were then used for comparison with s ediment grain size and glauconitic grain maturity to determine the extent to which the fossil assemblage descriptors of species richness and evenness vary with the changing environments up-section.


31 4. RESULTS Glauconitic Grain Maturity Many authors have described the limited range of physical environments in which glauconite will form (C loud 1955; Burst, 1958a,b; Hower 1961; Bentor and Kastner 1965; Porrenga, 1967; Odin and Matter 1981; Ireland et al. 1983). Glauconite formation generally occurs in the upper few meters of sediment in areas with s low sediment accumulation (Porrenga, 1967; Odin and Matt er, 1981) Gl a uconite formation is limited to these types of environments because of the gradual process in which glauconitic grains form (Odin and Matter, 1981). Over a period of 103 to 106 years, g l auconitic smectites and illites form in areas of reduced flow of interstitial waters such as within the structure of a c arbonate shell fragment or fecal pellet, within the chambers of a formaminiferal te st, or any number of other semi-isolated environments (Odin and Matter, 1981; Ireland et a!., 1983). Because of the limited ran ge of environments in which the g radual process of g l auconite f ormation c a n occur, the degree to which the g l auconite gra in s in a f ormation hav e completely formed (matured) c a n be d i re ctly re l ated to th e rate of sediment accumulation and se diment re-working whether by waves currents, or bioturbation (Ir e land et al., 1 983; Mallinson 1988).


32 Glauconitic grains recovered from the Gosport Sand were classified as either mature or immature, based upon their physical appearance. Color, shape, and luster are the primary characteristics upon which glauconitic grains are classified A direct correlation has been show to exist between these physical attributes and the chemical composition of the grains (Odin and Matter, 1981; Ireland 1983). Immature grains have an irregular to slightly rounded shape were light olive green to dusky green, and have a highly friable structure. Mature grains are rounded to globose, shiny, dark olive green to greenish black, and are tightly compacted. The patterns of glauconitic grain maturity from the two sites indicate similar histories of sedimentation rates and mixing. In both sections, the distribution s range from about 60% mature grains near the base of the Gosport Sand to nearly 95% mature grains near the top of the formation (Figures 9 and 10). The trends are not linear, but they display similar patterns in each section. Glauconitic maturity increases slig htly in three cycles, with reductions in glauconitic maturity occurring one-third and two-thirds of the way up-sec tion. The overall trend is one of increasing glauconitic maturity, suggesting that, as the Gosport Sand was deposited sediment accumulation rates gradually slowed


LS1 A LS2 A a.. LS3 L S 4 LS5 Q..l L S6 < 0 001= L S 7 0 c:c LS8 v LS9 v LS1 0 Figure 9. 0 33 MATURITY OF GLAUCONITIC GRAINS Little Stave Creek 3 0 m 10 20 30 40 50 60 70 80 90 100 PERCENT OF GLAUCONITIC GRAINS I MATURE GRAINS Glauconit i c Grain Maturity; Little Stave Creek. G l auconitic grains become more mature u p-section, s u ggesting reduced sedimentat ion r ates, which allowed g l auconitic grains to remain in the zone of formation for longer periods of time


1\ 1\ a_ ...J: <0 CIH-1-0 co v v CLA1 CLA2 CLA3 CLA4 CLA5 CLA6 CLA7 CLA8 CLA9 CLA10 CLA11 CLA12 CLA13 CLA14 CLA15 CLA16 CLA17 CLA18 CLA19 CLA20 CLA21 CLA22 CLA23 0 34 MATURITY OF GLAUCONITIC GRAINS Claiborne Bluff 1 0 20 30 40 50 60 70 80 90 1 00 PERCENT OF GLAUCONITIC GRAINS I MATURE GRAINS 6.9m Figure 10 Glauconitic G r ain M a turity; C l aiborne Bluff. A s in the Littl e S tave C reek sec tion, g l auconitic grains become more m ature up -sect ion D espite the expa nded sec tion (ove r 2x t h e thickness of the Littl e Stave Creek section), t h e pattern of g lauconitic maturity remains t h e same.


35 Sediment Characteristics The grain-size distribution of unprocessed samples from the Gosport Sand reflects both the terrigineous and the biogenic portions of the sediment. At both Little Stave Creek and at Claiborne Bluff the gravel percentages reach a maximum near the middle of the section (Figures 11 and 12). Sand percentages vary inversely to gravel percentages. Percent mud remains relatively constant, showing a slight increase up-section at each locality. The gravel-sized sediments are all shell material, while the sand fraction is approximately 1530 % biogenic The lack of evidence of post-mortem transport on the fossils (surfaces are generally pristine, except for clionid-type borings) suggests that the shells may have been rapidly buried. The percentages of sand and mud in the non-carbonate fraction of the sediment are fairly constant-approximately 85% and 15% at Little Stave Creek and 90 % and 10 % at Claiborne Bluff (Figures 13 and 14). The mean grain size of the non-carbonate fraction at Little Stave Creek is 2.9 to 1.9 phi, remaining very constant up section (Figure 15). At Claiborne Bluff, the mean grain size is fine sand (1.9 to 1.5 phi) and al s o remains fairly constant (Figure 16). Standard deviation of the sediment size, a measure of sorting is in the moderate to poorly sorted range throughout both sections, becoming slightly better sorted at the top of the Little Stave Creek section and slightly more poorly sorted at the top of the Claiborne Bluff section (Figures 15 and 16).


MUD, SAI\JD, GRAVEL, TOTA L CaC03 Raw sediment sample; Little Sta v e C ree k LS-1 LS2 LS-3 LS-4 w LS-5 _J o GRAVEL o S AI\JD 0 MUD CaC03 Q_ LS-6 3.0nl (/) LS-7 LS-8 LS-9 LS-10 0 1 0 20 30 40 50 60 7 0 80 90 1 00 PERCEI'H Figure 11. Mud, Sand and Grave l Percentages in Raw Samples; Little Stave Creek All of the g ravelsize d material is c a lcium carbonate (shell material). 36


MUD, SAND, GRAVEL, A I\JD TOTA L CaC03 Raw sediment sample; Claiborne Bluff CL-1 CL-2 CL-3 CL-4 CL-5 CL-6 CL-7 C L-8 CL-9 o GRA VEL o SAND <> lv1UD .t. 7. CaC03 CL-10 wCL-11 ciCL-1 2 2CL-1 3 Z?icL-1 4 6.9m CL-15 CL-16 CL-17 CL-18 C L-19 CL-20 C L-21 CL-22 CL-23 0 1 0 20 30 40 50 60 7 0 80 90 1 00 PERCE I'-ll Figure 12. Mud, Sand and Gravel Percentages in R aw Sample s; Claiborne Bluff. All of the gravel-sized m aterial is calcium carbonate (shell material). 37


38 S A N D AND tv1UD N on -carbonate fraction; Little Sta v e C ree k LS-1 L S 2 o S A I'-JD 0 MUD LS3 LS-4 w L S-5 _J Q_ 3.0 nl 2 LS-6 <( if) LS-7 LS8 LS-9 L S 1 0 L-..1.-..:!\::.L, 0 1 0 2 0 3 0 40 5 0 6 0 7 0 8 0 90 1 00 PERCEI\lT Figure 13. Mud and Sand Percentages of Non carbonate Sed imen ts; Little Stave Creek. The lack of significan t variation up-sec t ion suggests that there were few, if any changes in the hydraulic reg i me or the sou r ce of the sediment during the period of the deposition of the Gosport Sand. Th i s does not preclude changes in the rate of sedimentation as the overall sediment supp l y may have been reduced at the source whi l e retaining similar grain-si z e patterns.


S A I'-10 lv1U D Non-carbona t e fractio n ; C l a i b orne Bluff CL-1 CL 2 C L-3 CL-4 C L-5 C L-6 CL-7 C L 8 CL-9 CL-10 w CL-11 2 <7icL-1 4 CL-15 CL-16 C L-17 CL-1 8 C L-19 C L -20 C L-21 CL-22 C L 2 3 0 1 0 2 0 30 4 0 50 60 7 0 8 0 90 1 00 PERC E I\JT 0 0 6. 9 m S A N D MUD 39 F igur e 14. Mud and S a nd P e rcent ages of Non-carbon a te S edime nt s ; : C laiborn e Bluf f Alth o u g h s lightl y s andier th a n th e se d i m e nt s f r om Littl e Sta ve C r e ek th e sa m p l es fr o m Cl a iborn e Blu f f display a s imilar l ac k of var i a tion in gra in s i ze.


lvlEA I\l A I\JD S T A I\JDARD DEV I ATION I 'Jon -carbonate fraction; Little Stav e Creek LS-1 L S 2 LS-3 LS-4 w LS-5 ....J D.... LS-6 Vl LS-7 LS-8 LS-9 LS-10 0 1 2 3 lv1EA N ( phi ) I STD. D E V 40 0 MEA N 0 STD. DEV 3.0m Figure 15. Mean Grain Size and Standard Deviation of Non-carbonate Sediments; Little Stave Creek. The lack of variation in mean grain size and in sorting (standard deviation) suggests that depositional energy (currents waves) remained relatively stable during the deposition of the Gosport Sand.


S T AI'-JDARD DEV I ATION Non -carbonate fraction; Claiborne Bluff CL-1 CL-2 CL-3 CL-4 CL-5 CL-6 CL-7 CL-8 CL-9 CL-10 wCL-11 0:CL-12 2CL-13 4 CL-15 CL-16 CL-17 CL-18 CL-19 CL-20 CL-21 CL-22 C L-2 3 _.__ ___,jj; ____...--'":l--.L-----'------l 0 1 2 3 (phi) / STD. DEV o MEA N <> STD. DE V 6.9m Figure 16. Mean Grain Size and Standard Deviation of Non-carbonate Sediments; Claiborne Bluff Wave and current energy remained relatively constant, as at Little Stave Creek. 41


42 The grain-size distribution of the carbonate fraction is perhaps more indicative of the changes that took place after the deposition of the Gosport Sand. Both studied sections show a carbonate content of approximately 60% in the lower half of the section, which gradually declined to about 20 % toward the top (Figure 17 and 18). The more carbonate basal sediments are coincident with elevated percentages of carbonate gravel while carbonate sand and mud percentages increase toward the top of the section where overall carbonate levels drop The decline in carbonate gravel percentages toward the top of the section is likely due to the increased weathering to which the fossils have been subjected in the upper layers of the formation. Due to this increased degradation the validity of the grain-size distribution of the carbonate fraction of the upper samples is questionable. Microscopic examination of the sand and mud fraction of the raw sediment throughout the section revealed no significant component of whole molluscan fossils in these size ranges, indicating that the change in the carbonate gravel/sand/mud percentages was not due to a change in the actual average size of the fossils but was due more to a reduction in size of the shell fragments as a result of the increased weathering.


T O T P L CaC03 COI\JTENT, lvlUD, SAND, AND GRAVEL 7. Carbonat e fracti o n; Little Stav e C ree k LS-1 LS-2 LS-3 -LS-4 I w LS-5 I I _J i Q_ I 2 LS-6 I <( (/) I LS-7 LS-8 I L S 9 -LS-1 0 0 1 0 20 30 40 50 60 70 80 90 1 00 PERCEI\JT o GRAVEL o S AND 0 MUD 7. CaC03 3.0m 43 Figure 17. Sediment Carbonate Content and Mud, Sand and Gravel Percentages of Carbonate Sediments; Little Stave Creek. Total carbonate content and percentage of carbonate sediment which is gr a vel-sized decrease towards the top of the section where weathering has reduced many of the shell fragments to sand or mud size.


TOTA L CaC03 COf\JTEI\JT, lv1UD, AND GRAVEL 7. CL-1 CL2 CL-3 CL-4 C L-5 CL-6 CL-7 CL-8 CL9 CL-10 wCL-1 1 o=cL-12 2CL-1 3

45 Faunal Diversity and Species Occurrence At the onset of the fossil sorting and identification it became obvious that all of the samples collected would not be suitable for the faunal study. Samples retrieved from the upper 1.5 m of the Gosport Sand at Claiborne Bluff, while shelly, were so weathered that few intact specimens could be recovered from the dried sample. For this reason, only samples CL-6 through CL-23 were utilized in this portion of the study. The rarefaction graphs (Figure 7) for each sample fit the log-normal distribution, suggesting that the samples were large enough to overcome any bias introduced by unequal sample sizes (Magurran, 1988). In all samples only a few species were represented by a large number of individuals, while most species were sparsely represented While this initial analysis suggests that all samples were of a size sufficient to overcome sampling bias it is advisable to examine the relationship between sample size and the occurrence of common species and individuals in each sample in order to qualify changes in the indices of species richness and evenness (Magurran 1988). The distributions and abundances of all species recovered from the Gosport Sand, as presented in Appendix 2, are not meaningful unless the rare species are removed from the stratigraphic database This is because species must be common in order to be biostratigraphically significant (Koch, 1973; Cobabe and Allmon, 1993). Rank/abundance graphs for each sample (see example Figure 8) were prepared, and 43 species were


46 classified as being common in at least 1 sample The stratigraphic occurrences of the s e species are depicted in Figure 19. No distinct faunal turnovers are apparent; changes in the fossil assemblage are gradual, and species cooccurrence patterns range up section. From the distribution of common species through the section plots of the number of common individuals and species in each sample were constructed (Figure 20), with sample sizes plotted for comparison These values were used to calculate the diversity indices of DMg (species richness) and 1/d (species evennes s ) (Figure 20). Species richness measures the relative number of common species in a given sample while evenness is a measure of the relative abundance of each species in the sample (May, 1975; Magurran, 1988) Differing sample sizes do not appear to affect the patterns of total or species abundance, or species richness and evenness. The v alues of the indices of species richness and evenness follow similar patterns stratigraphically A slightly increasing trend in the values of both indices i s obs e rved in the upper two-thirds of the studied portion of the section, which represents the middle one-half of the entire exposure (the upper five samples were not included in the faunal analysis due to poor fossil preserv a tion).


300 individu a l s 47 = CL-6 CL7 CL-8 CL9 CL-10 CL-11 CL-12 CL-13 CL-14 CL-15 C L-16 CL-17 CL-18 CL-19 CL-20 CL-21 CL-22 CL-23 I I I I tilt. 'I I 111 I I I I I I I I I I I I I' I I I I I I I 2 3 -4 5 6 7 8 9 191, 121:3 '" 1516171 8 192921 22232<425262729293931 Venericardia (Pleuromeris) parva Lea 2 Linga (Cavi lin ga) pomilia (ConT"dd) 3 Turritella carinat a !.Lea 4 Retusa (Cylichni na) galba (Conrad ) 5 Venericardia (Venericardia) r o tund a Lea 6 Glycymeris trigo n ella (Conrad) 7 Mesalia vet usta (Conrad) 8 Gratelupia (Cytheriops is) hydana (Conrad) 9 Callista (Costacallista) aequorca (Conrad) 10 Diplodonta (Diplodonta) ungulina (Conrad) 11 Spis ul a parilis (Conrad) 12 Cacstocorbu la murchis onii (Lea) 13 Caryocorbula alabamicnsis (Lea) 14 Glycymcris trigonella minor (Lea) 15 Ca ll ista ( Callista) pcrovata (Conrad) 16 Caryoco rbula alabami cns i s gregori oi (Cossmann) 17 Vcncricardia (Ciaibo rnicardia ) a lticosta ta (Conrad) 18 Crepidula Jirata Conrad } 9 F.gcrel l a limatul: : 20 Linga (Cavi linga) pomilia alveata (Conrad) 21 Tellina (Eurytellina) papy ria Conrad 22 Caiyptraphorus ve latus (Conrad) 23 24 25 26 27 28 29 30 3 1 3 2 33 34 35 36 37 38 39 40 .1] 42 43 Glyc ymeris idonea (Conrad) J\'ucula (Nu c ula) magnifi ca Conrad Parmicorbul a gibbosa (Lea) Platytrochus s t o k esi (Lea) G l ycy mcris s taminea (Conrad) N ucu la (Nu cula) ovula Lea Nuculana coclata (Conrad) Agaronia a labamensi s (Conrad) Astarte proruta Conrad Bullata se men ( !.Lea) Callis t a (Callista ) perova t a lisbonensi s { Harris ) Callista (Callista) perovat a s ubvitrca (de Gregorio ) Crepidula dumosa Conrad Dcnt alium minu t istriatum Gabb Epilu c ina r otunda (Lea) Lirodisc us tcllin oidcs (Conrad) Ncvcri la limula (Conrad) Pilar (Pilar) nuttali (Conrad) Turritclla apita de Gregorio T u rritclla obruta Conrad Figure 19. Vertical Distributions and Ab und ances of Species th a t are Common in at Least One Sample. The width of the scale bar repre se nts 300 individuals Ranking of the spec i es is in decreasing order of number of samples in which the species a re common.


2 Figure 20. SAMPLE SIZE Claibo rn e Bluff SAMPLE SIZE ( cnU ) SAMPLE SIZE C lai borne Blu ff SAMPLE SlZE (null 1 0 0 48 TOTAL ABUNDANCE SPECIES ABUNDANCE ( C ommo n F o ssils ) (Common Fossils ) 400 600 1200 1600 NUMBER OP COMMO N I N'DrvtDliAU SPECIES RICHNESS (Common F ossils ) 0 5 1 0 15 20 NUMBER OP COMMO N SPEClES SPECIES EVENESS (Common Fossils ) 5 4 m 25 5 .4m Sample Sizes Number of Common Individu als per Sample Number of Common Species per Sample Specie s Richness, and Evenness; Claiborne Bluff.


49 5. DISCUSSION Background Work Previous paleoenvironmental interpretations of the Gosport Sand have been few compared to the number of taxonomic studies. Cooke (1939) described the Gosport fauna as shallow water because it contains a high percentage of littoral species. He considered the Gosport Sand and Moody's Branch Formation as having been deposited by a transgressive sea Gardner (1957) interpreted the Gosport Sand as representing a beach deposit, similar to Sanibel Island Florida She concluded that the organisms lived nearshore at a maximum depth of 120 feet deep and in temperatures slightly warmer than that of the present-day Gulf of Mexico, and were transported landward prior to their deposition. Toulmin (1977) based on the fossil assemblage and the presence of carbonaceous leaf-bearing clays interfingering with the Gosport concluded that the Gosport represents a nearshore environment. Swindel (1986) identified the deposit as a transported assemblage deposited on a beach. Allmon (1988), in his extensive summary of work on the Gulf and Atlantic Coastal Plains, described the paleoenvironment of the Gosport as 0-20 meters in depth, tropical (mean surface water temperature >20 degrees C), shelly sand substrate, normal salinity (33-37 o/oo), low turbidity,


50 moderate-to-very agitated turbulence, normal oxygenation and in the inner shelf ecological zone. Minor changes in the chemical and physical environment can effect changes in the structure of a molluscan community without causing regional mass extinctions (Hansen, 1987). Such changes may be recorded in resultant fossil deposits as variations in the diversity and abundance of individual fossil species (May, 1975) The physical and biological characteristics of the depositional environment can alter the degree to which the fossil assemblage accurately represents the actual living community from which it was derived. Taphonomic processes such as these overprint the original patterns of diversity and species turnover by breaking shells and mixing originally distinct communities (Staff et al. 1988). The resultant skeletal deposit is frequently not a reliable representative of the original biotic community To assess the degree to which taphonomic forces have altered the fossil record and in an attempt to correlate environmental fluctuations with changes in community structure patterns of sediment grain size and composition were examined as proxies for changes in the physical environment. While many factors (sediment and water chemistry food supply, current direction, types of currents) control the make-up of benthic invertebrate communities (Rhoads, 1974), sediment grain size is the most easily measured in the stratigraphic record A detailed study of grain size and sorting trends gives some indication of the stability in the physical environment during the deposition o f the Gosport Sand. Changes


51 in the sedimentary regime (if any) occurring in concert with changes in the faunal constituents would support the hypothesis that changes in the phy s ical environment were the cause of the changes in the fauna If a range in grain size were not available, however, it would not be possible to observe these changes Grain size of the carbonate constituents of the sediment are also indicative of phys i cal processes; skeletal fragments break dissolve and become more abraded. The more physically degraded a shell fragment is, the more physical and chemical activity it has been subjected to after its death ; shell degradation is often used to characterize the level of physical and chemical activity in the environment of deposition (Staff and Powell 1990). Cummins et al. (1986) showed that the size-frequency distribution of a death assemblage is not only often signific a ntly different from that of the original living community, but is also a direct result of the unique taphonomic processes to which the death assemblage has been subjected The size of the carbonate fragments therefore is in part indicative of the degree to which the assemblage has been dissolved or reworked (through the action of burrowers currents, or waves), and therefore reflects the degree to which the diversity and first/last appearance patterns are representative of the original life assemblage (Staff and Powell, 1990). One other way of analyzing the sedimentological and taphonomic changes of the Gosport Sand is by examining the stratigraphic pattern of glauconitic maturity. Because the degree of maturity of glauconitic grains is related to how long they remain in the zone of glauconite accumulation, both burial rates and the


52 degree to which the shallow sediments are reworked by currents waves and biological activity will affect the maturity of the glauconitic grains as well as the degree of degradation of the carbonate grains (Odin and Matter 1981; Mallinson 1988; Staff and Powell, 1990). Implications of Sediment Characteristics The grain-size trends of the non-carbonate sediments (Figures 13 16) indicate little change in the hydraulic regime throughout the section The calculated suspension velocities of sand grains in the range of diameters found in the Gosport Sand (1.5 to 2.5 phi) range from 20 to 30 em/sec (All e n 1970) suggesting that there was little overall change in the hydraulic regime throughout the time of deposition of the Gosport Sand. It does not necessarily mean that sedimentation rates remained static however. The quality of preservation of the shells decreases toward the top of the section, resulting in a decrease in carbonat e grain size as the shells are more easily broken This suggests that the rate of shell burial decreased as the Gosport Sand was deposited. Decreasing rates of sedimentation may have allowed increased shell dissolution and/ or breakage Anothe r possibility is that water depth increased, preventing the average storm wave-base from affecting the seafloor and thus decreasing any possible storm related mixing of the upper layers of the sediment. No sedimentary structures were observed, however, that would suggest that there was storm-related mixing in


53 the Gosport Sand. Post-depositional dissolution also contributed to shell degradation in the Gosport Sand, possibly facilitated by clionid-type borings One possibility is that shells that were exposed on the seafloor for a longer period of time were more extensively bored by clionid-type organisms, creating a greater surface area (in the form of pore spaces) for post-depositional dissolution to act upon. Heavily bored shells that would fit this model are generally restricted to large epifaunal gastropods. In general, the shells near the base of the Gosport Sand were more broken, while those nearer the top of the section were more highly dissolved. A possible explanation for the observed trends in shell deterioration is that lower sedimentation rates during the upper, condensed portion of the Gosport Sand resulted in the fossils remaining in the upper few meters of the sediment for a longer period of time. This zone, termed the taphonornically active zone (T AZ) by Staff et al. (1988), is where the majority of fossil destruction takes place via bioturbation (Staff and Powell 1990). If the effective depth of bioturbation remained constant while the rate of sedimentation decreased, the result would be more thorough bioturbation and consequently more shell breakage (and therefore smaller sized carbonate grains) due to increased bioerosion. Dissolution of shells would have increased as well; a longer exposure on the seafloor allows more of the shell material to be dissolved by seawater (Davies et al. 1989). Thorough bioturbation (destruction of virtually all sedimentary laminae) can occur at depths of up to at least 180 em below the sediment-seawater


54 interface in shallow continental shelf areas (Rhodes, 1974). The virtual lack of sedimentation shelfward of the inner shoreface (Holmes and Martin, 1977), contributes to the ease of loss of the geologic record of any but the longest-lasting environmental trends. In the estimated 1.5 million years that it took to deposit the 3-to 12-meter thick Gosport Sand (Hazel et al., 1984) the effective amount of time represented by the sediments affected by bioturbation averaged 225,000 to 900,000 years. The actual amount of the geologic record so affected was probably much longer in the later stages of deposition because sedimentation rates slowed during the sea-level highstand. The most important result of slowed sedimentation during the latter stages of the deposition of the Gosport Sand is the condensation of the stratigraphic record. Slower sedimentation rates resulted in increased apparent species diversity in the samples near the top of the section; becaus e it took lon ger to deposit a standard thickness of sediment near the top of the Gosport Sand than at the bottom the number of species in a given interval increased If the thickness of the sample intervals were decreased in the upper l ayers (slower sedimentation) keeping the amount of time represented by an individual sample equal, the resultant diversity patterns would likely be more representative of the actual diversity of the living community.


55 Interpretations of Glauconitic Maturity Glauconite is a common constituent of Gulf Coastal Plain sediments. Despite this, it has been infrequently utilized in paleoenvironmental interpretations. The actual environmental requirements for glauconite formation are not well understood (Bentor and Kastner, 1965), and the composition of glauconitic grains Al, Fe2+, Mg)-2010 (OH)2K(x+y) -is highly variable, making it difficult to relate glauconite composition to specific paleoenvironmental conditions or to assign reliable dates to them (Odin and Matter, 1981). The variable composition is due to the process of glauconitization, a two-stage precipitation-dissolution-recrystallization process (Odin and Matter, 1981; Ir e land et al., 1983; Borhold and Giresse 1985). The gradational process of g lauconite formation does, however, allow one to make relative comparisons regarding the l e ngth of time that the g l auconitic grain remained in the narrow zone of glauconite formation Mallinson (1988) summarized the process in which the substrate (typically a mollusk shell fragment or foraminifera) hosts the initial growth of glauconitic smectite crystals within microscopic fractures and pore spaces, either directly in the framework of the shell (in the case of a mollusk fragment) or within the chambers of a gastropod or foraminifera. Subsequently, the grains are dissolved and replaced by nascent glauconitic smect ites. Following this stage potassium is progressively incorporated into the interlayer positions of the g l auconitic smectite


56 as water is simultaneously lost. In this fashion the structure is driven towards an illitic arrangement and reflects a highly evolved glauconitic grain ( Ireland et al. 1983) This progressive transformation from glauconitic smectite to a glauconitic illite will only take place in a limited depositional envirorunent: at or near the sediment/seawater interface in a transition zone between oxi di zing conditions above and reducing conditions below where ferrous and ferric iron coexist (Ireland et al., 1983). The residence time of an evo l ving glauconitic grain within this zone of formation is directly related to the degree of maturity of the r esultant grain. Odin and Matter (1981) reported that semi rounded, yellow green to olive green glauconitic smectite grains (immature) may form from an altered carbonate framework or fecal pellet in a few thousand years. Green, cracked grains may develop after approximately 104 years. Mature g l auconitic illite requires 105 to 106 years and is typified by well-rounded, dark green to nearl y black grains. At any time in this process, glauconitization may be halted by burial, removing the grain from the crucial zone of formation (Odin and Letolle, 1980). By utilizing this characteristic, Mallinson (1988) showed that a relative estimate of sedimentation rates may be deduced from the relative maturity of grains. Bec ause such long residence times in the zone of formation are required se diment ation rates must be slow enough to allow the sediment/water interface and therefore the zone of co-existence of ferric and ferrous iron to remain nearly static. This provides sufficien t time under favorable condit i ons for glauconitization to take


57 place. Rapid sedimentation would continually shift this zone upward causing iron that might have been taken up by glauconitization to be reduced and incorporated into sulfides in the zone of sulfide reduction (Burst, 1958a; Bentor and Kastner, 1965). Slow sedimentation rates are not the only controlling factor on glauconitization. Also required are suitable substrate grains, in terms of size and mineralogy, and an abundant supply of iron (Bornhold and Giresse, 1985) The substrate grain must be between 0 1 and 2.0 mm, because the restriction of the movement of pore water is required for glauconitization (Odin and Matter, 1981). Porous carbonate grains fecal pellets, gastropod shells, and foraminiferal tests are ideal for such circumstances, as they provide a micro-environment in which glauconitization might occur. The necessity of abundant iron is perhaps the most intriguing requirement for glauconitization Open marine waters contain levels of iron as low as 0.01 ppm (Horne, 1969) The majority of continent-derived iron is trapped in nearshore environments such as estuaries and lagoons (Fairbridge, 1967). In modern marine environments, however, glauconitic minerals form in areas where the seafloor is between 60 and 100 meters deep (Cloud, 1955; Bo rnho ld and Giresse, 1985). This apparent contradiction was possibly resolved by Odin and Matter (1981) in their suggestion that eustatic seale vel changes might be responsible for introducing iron into deeper water. Iron-rich, shallow-water sediments deposited during a sea-level lowstand would be inundated during a


58 transgression Sedimentation rates would also diminish with increasing transgression as sediment sources (river mouths) were shi f ted t oward the continent with the rising transgression. With sufficient bioturbation and a reduction in sedimentation, these iron-rich sediments may then be made availab l e for glauconitization (Odin and Matter, 1981; Mallinson 1988). Odin and Matter (1981) divided the maturational stages of glauconitic grain development into four categories: nascent (Type 1), little-evolved (Type 2), evolved (Type 3), and highly evolved (Type 4). These grain-types are recognizable to the trained eye under a standard binocular microscope or a polarizing petrographic microscope. Type 1 grains appear as slightly altered carbonate grains, usually still recognizable as the original grain (shell fragment fecal pellet formaminifera test etc.) with a surficial coating of pale-green glauconitic smectite. Type 2 grains have been partially replaced by glauconitic smectites, with pore spaces penetrating the grain infilled with light-green to moderate-green glauconitic zones. Type 3 grains have been almost completely replaced with smectites re-arranging into less expandable illite; surficial crack s appear due to faster crystal growth inside the light to mode r ate green grain than on the surface. Type 4 grains are fully developed illite, moderate green to du sky green are round to subround, irregular to multi-lobed and are commonly polished with surface fissures. It must be kept in mind that these divisions are artificial as glauconitization occurs along a continuous spectrum. The division s are convenient, though, for characterizing the degree o f matur ity o f the


59 glauconitic components of a sediment sample. In practice it was extremely difficult to discern between Odin and Matter's four types of glauconitic grains, so in this study the glauconitic grains are classified as either mature (Types 3 and 4 ) or immature (Types 1 and 2) Photomicrographs of examples of mature and immature grains under a polarizing microscope are depicted in Figure 21. In highly bioturbated sediments, burrows which remain open for any length of time will allow interstitial circulation to greater depths than adjacent sediments (Rhoads, 1974). This situation, in which the sediments are unevenly exposed to fresh seawater, can result in a range of the degree of glauconitic grain maturation; the boundary between oxidizing and reducing environments varies vertically as well as horizontally through time as burrows infill and are replaced by new burrows (Cloud, 1955). Unlike grain-size distribution which is not as easily alte r ed by bioturbation, glauconite formation is controlled by factors that can be widely variable in a short distance namely the chemical boundaries of the micro environments in which the glauconite can form. For this reason glauconitic maturity can only be used as a basis for comparison of rates of bioturbation and environmental changes between successive stratigraphic layers not as an empirical indicator of environmental conditions in an isolated sample (Ireland et al. 1983). In the samples examined in this study glauconitic maturity increased stratigraphically upward, from near 50 % mature grains at the base of the Gosport


60 Figure 21. Photomicrographs (Crossed Polars) of Mature and Immature G l auconitic Grains. The top picture is of an immature glauconi tic grain in which glauconitic smectites fill pores in a shell fragment. The bottom picture is of mature glauconitic grains th at have been complete l y rep l aced by glauconite, becoming more dense and rounded as original substrate grain was completely replaced. The width of the scale bar is equal to 2 nun.


61 Sand to almost 95% mature grains at the top of the formation (Figures 9 and 10). At Claiborne Bluff (the more expanded section) the maturity increases upward for the first 1.2 meters, followed by a slight drop in maturity. From this point on, the maturity increases to the maximum at 2 6 meters, where it decreases slightly before increasing again toward the top of the section. The general trend of increasing glauconitic maturity indicates an overall slowing of sedimentation rates, possibly accompanied by increased sediment mixing. No radical changes in the pattern of glauconitization were noted, suggesting that changes in sedimentation rates were gradual. As sedimentation rates slowed the amount of time represented by successive, equal thickness layers of sediment increased. Sedimentation Rates The patterns of sediment grain size and glauconite maturity observed in the Gosport Sand suggest that there was a trend for sedimentation rates to decrease throughout the deposition of the formation. This agrees with Mancini and Tew (1991) in their placement of the Gosport Sand in the framework of relative sea levels in the eastern Gulf Coastal Plain (Figure 22). Following the Lisbon-Gosport Sand regression and associated unconformity the Gosport Sand was deposited relatively rapidly during the initial transgression. Sedimentation rates gradually slowed resulting in the condensed section found in the upper few meters of the Gosport Sand.


62 Super--8 Relative Chang111in Sequence >'1atUliDn< Aelatlvv I Lanclwllc:"lal en:.,.td Lithology Ulhosttallgraphy Group Fotan"'in;f., .. Love! Chanf.n cycle t'i CofTl'O'ltniS Zone don "' \ sands & clays H i !t!olllnd Red B'ufflforH I Hit Td ,.; days. & Kmestones CondenMd Seed"" Red Btuii/Bu"'!>"'M lU clays Tranagresaive ShuboJIII Gr. CMTN:CU ffKi .... Plic:hota She l f Marain Cocoa \ Higholllnd Net" Twia!WOod h P. r N days Jackson I.Z. ,.; CondonMd Section Crook w & glauconillc sands .... Transg&Ne Moodys Ihnen sands, clays & lignite I I T tOhri ,.; w COtldonMd Sec don Gosport I Z .... & glautonlllc oands Transgreu.ive \ sands & shales Highslllnd Codaciulko II c N Zilpha IUbconpJobolo w COtldanMd Sec ion C .RZ. .... masts & glaucon itic sands TtanSOtuN w..,.. ----------\ ... ,,, sands & days Hig hllllnd Noohobo N & sills COtldanMd S.Cion Buic:City j H .,_gOMI'IIiS w & g uconlllc: oands .... 'il sands Lowsta nd MO!idlon ... ---------II sands, silts & clays Highstand Hatcl1otigboe Tb COtldonMd Sec ion Wolcox At ubboiMe I.Z. & glautonillc sands Transgr.uive Buhi I s I I I s I s Type 1 unconbmuty I I I I I I I I I Type 2 u nconbmry Figure 22., Generalized Rel ative Sea Level Cha nge During the Middle to L ate Eocene in the Eastern Gulf Coastal Plain (Afte r Mancini an d Tew, 1991). I


63 This slowing of the rate of transgression resulted in a decrease in the degree to which the fossil record accurately represents the original death assemblage. The sediments preserved in the lower section of the Gosport Sand were buried without significant condensation, and more accurately represent the original fauna. In the upper, condensed section, the fossil record is less trustworthy, because more time is represented by intervals of the same thickness as those at the bottom of the section. The result is that samples in the condensed section contain fossils in greater abundances than occurred in the living communities. Faunal Turnover and Diversity Trends Cobabe and Allmon (1993) showed that the sample size necessary to adequately sample the moderately common and rare species in the Gosport Sand would be prohibitively large. Because of this, only common species can be used in the calculation of species richness and evenness and in the tabulation of occurrence and abundance patterns. The indices of species richness and evenness remain relatively constant upward in the Gosport (Figure 20), but the extreme variability in the vertical distributions and abundances of the common species (Figure 19) suggest that the successive communities were comprised of highly variable faunal assemblages. The changes in community structure were evidently driven by forces that were not strong enough to effect a change in overall


64 community diversity. Such forces are likely to be intrinsic to the community, such as competition and niche partitioning (Korniker and Odum, 1967). No total overturning of the fossil assemblage was observed, suggesting that no widespread extinctions occurred during the deposition of the Gosport Sand. Rather, as a dominant species waned from its peak occurrence, another would increase in dominance to fill the void in the community structure. The increased abundance in the upper section of the Gosport Sand of six of the seven most common species can be attributed to increased time-averaging of successive communities. An analogous situation is if one were to remove 80 % of the flakes from a box of raisin cereal. The number of raisins in the box would remain the same, but the proportion of the raisin s among the flakes would increase. The compression of the record of increased lengths of time caused more fossils to be accumulated over a given vertical distance near the top of the section as compared to the bottom The condensation of the temporal record resulted in a further increase in apparent abundance while diversity and evenness remained stable. Had the Gosport Sand been a result of the accumulation of the fossils from a series of highly diverse, long ranging molluscan communities the composition of the communities should have remained relatively stable through time (Slobodkin and Sanders 1969) While this scenario might also result in little variation in the indices of species richness and evenness, the relative abundances of the common species should also remain fairly constant. This, however, is not the case in the


65 Gosport Sand (Figures 19 and 20). Little correlation exists in the co-occurrence of the common species in the Gosport Sand; while species richness and evenness may have remained stable, the abundances of individual species vary independently up-section. This pattern supports the hypothesis that the molluscan communities represented by the fossils of the Gosport Sand changed in structure through time. The extreme diversity displayed in the Gosport Sand is therefore artificially high due to the time averaging of fossils from successive depositional layers.


66 6. CONCLUSIONS The Gosport Sand was deposited in a relatively shallow ( <30 meters) shoreface environment (Cooke 1939; Toulmin 1977; Allmon, 1988). During the initial transgression, sedimentation rates were at their maximum for the formation, so the shells were buried relatively rapidly, resulting in a well preserved assemblage with minimal breakage. As transgression slowed, sand influx was diminished, allowing the fossils to remain near the sediment-seawater interface for a longer period of time, during which the y were su bjected to more phy s ical and chemical degradation. These deposits are characterized by a lower total carbonate percentage and increa se d levels of carbonate sand and mud, as compared to sediments in the lower Gosport Sand. The effects that the se sedimentation rates had on the fossil record include a condensing of fauna from successive original communities to such a degree t ha t discrete time intervals cannot be confidently identified. The changes that occurred in the sedimentological regime were gradual ones, however, allowing for trends in sedimentation patterns to be compared with changes in faunal patterns. The more rapid sedimentation rates at the beginning of the deposition of the Gosport Sand resulted in better preservation of the fossils and a more reliable repre sen tation of the original community near the ba se of t he Gosport as


67 compared to the top. Some faunal mixing may have occurred at least in the lower layers, however, as burrows from the lowermost beds of the Gosport Sand commonly channel shells from the Gosport Sand at least 30 em down into the Lisbon Formation. The degree of condensation of the fossils in successive layers increased through time as the Gosport Sand was deposited. Of primary importance in this study is the lack of sedimentary indicators of rapid environmental change. This is not to say that such changes did not occur, only that if they did occur, they have been obscured by the mixing of successive layers of sediments. The resultant pattern of gradual change throughout the section suggests that any environmental changes that may have occurred were not concentrated during any discrete interval, but rather were spread out relatively evenly through time. The trend of these changes was directional ; rapid deposition and carbonate accumulation during the early inundation of the shelf, with bioturbation having a marginal effect, followed by gradually slower rates of sediment accumulation and poorer preservation of shell material in the condensed section, during which bioturbation was the primary agent responsible for faunal mixing. The resultant effect upon the skeletal deposit is that the record of community succession at this exposure is likely to be more trustworthy near the base of the section than toward the top. Apparent species abundances in the condensed section are likely higher than what would be expected in beds that were not time-condensed.


68 The extraordinarily high molluscan diversity observed in the Gosport Sand therefore, is due primarily to the time-averaging of successive communities and not to changing environmental conditions; that is to say that post-depositional sediment condensation resulted in artificially high abundances of all species. These high abundances increase the chance that species will be recovered that were re latively uncommon in the living community. Community structure indices (species richness and evenness) change little up-section suggesting that changing environments of deposition did not alter the diversity of the living communities.


69 REFERENCES Allen, J.R.L. (1970). Physical Processes of Sedimentation New York: American Elsevier. Allmon, W.D. (1988). Evolution and environment in Turritelline gastropods (Mesogastropoda, Turritellidae), Lower Tertiary of the U.S. Gulf and Atlantic coastal plains. Unpublished doctoral dissertation Harvard University. Allmon, W D. (1989) Paleontological completeness of the record of Lower Tertiary mollusks, U.S. Gulf and Atlantic Coastal Plains: Implications for phylogenetic studies. Historical 141-158. Bandy, O.L. (1949). Eocene and Oligocene foraminifera from Little Stave Creek Clarke County, Alabama. Bulletins of American Paleontology, 32, 210 Bentor Y.K., and Kastner, M. (1965). Notes on the mineralogy and origin of glauconite. Journal of Sedimentary Petrology, 35, 155-166 Berggren, W.A. Kent D V., Flynn J.J., and Van Couvering J.A. (1985). Cenozoic geochronology Geological Society of America Bulletin 96, 14071418. Blanpied, B.W and Hazzard, R.T. (1938) Correlation of Cockfield and Gosport Formations, Eastern Mississippi and Western Alabama. AAPG Bulletin 32, 309-314. Bornhauser M. (1947). Marine sedimentary cycles of the Tertiary in the Mississippi Embayment and central Gulf Coast area. Bulletin of the American Association of Petroleum Geologists, 31, 698-712 Bornhold, B.D and Giresse, P. (1985). Glauconitic sediments on the continental shelf off Vancouver Island British Columbia, Canada Journal of Sedimentary Petrology, 55, 653-664. Burst, J F. (1958a). "Glauconite" pellets: Their mineral nature and applications to stratigraphic interpretations. Bulletin of The American Association of Petroleum Geologists 42 310-327.


70 Burst, J F (1958b ). Mineral heterogeneity in "glauconite" pellets. The Americ a n Mineralogist 43, 481-497. Buzas M.A., C.F. Koch, S.J Culver, and N F Sohl, (1982). On the distribution of species occurrence. Paleobiology _B, 143-150 Bybell, L.M (1975) Middle Eocene calcareous nannofossils at Little Stave Creek Alabama. Tulane Studies in Geology and Paleontology .ll, 177-254 Cloud P.E., Jr. (1955). Physical limits of glauconite formation. Bulletin of The American Association of Petroleum Geologists 39 484-492 Cobabe, E.A. and Allmon W.D (1993) (submitted) Effects of sampling on paleoecologic and taphonomic analyses in high diversity fossil accumulations: An example f rom the Gosport Sand Middle Eocene Alabama. Paleobiology. Cooke C.W (1939) The equivalence of the Gosport S a nd to the Moody's M a rl. Journal of Paleontology 13, 337-340. Cummins, H. E.N. Powell R.J Stanton, Jr., and G Staff (1986) The s iz e frequency distribution in palaeoecology : Effects of taphonomic processes during formation of molluscan death assemblages in Texas bays Palaeontology, 29, 495-518. Davies, D J. Powell, E.N. and Stanton R.J Jr. (1989). Relative rate s of shell dissolution and net sediment accumulation -a commentary : Can sh e ll b e d s form by the gradual accumulation of biogenic debris on the sea floor? Lethaia 22 207-212. Davies D.K., and Etheridge F.G. (1971) The Claiborne group of central Texas: A record of Middle Eocene marine and coastal plain depo s ition Transactions-Gulf Coast Association of Geological Societies 21, 115-124. Dockery D T. III ( 1 984) Crisis events for Paleogene mollusc a n fauna s in the southeastern United States. Mississippi Geology, _5, 1-6. Dockery, D T., III (1986). Punctuated succession of Paleo g ene mollusk s in the northern Gulf Coastal Pla i n Palaios, l 582-589 Dockery D T. III (1987). Eocene-Oligocene molluscan extinctions: Commentary Palaios _2, 620-621.


71 Fairbridge, R.W. (1967). Phases of diagenesis and authigenesis in Larsen, G and Chilingar, G.V. (eds.), Diagenesis in Sediments Amsterdam: Elsevier. Galloway, W.E. (1989) Genetic stratigraphic sequences in basin analysis II: Application to the Northwest Gulf of Mexico Cenozoic basin. The American Association of Petroleum Geologists Bulletin, 73, 143-154. Gardner, J.A. (1957) Little Stave Creek, Alabama-A paleoecologic study in Treatise on Marine Ecology and Paleoecology, Geological Society of America Memoir 67, _2, Paleoecology. Golden, J. (1969). Paleoecologic interpretations of Tertiary sediments based on the foraminifera. Unpublished master's thesis, Washington University Gross, M.G. (1982) Oceanography: A View of the Earth. Englewood Cliffs NJ: Prentice-Hall. Hansen, T.A. (1987). Extinction of late Eocene to Oligocene mollusks: Relationship to shelf area, temperature changes, and impact events. Palaios, _2, 69-75. Hardenbol, J., and Berggren, W .A. (1978). A new Paleogene numerical time scale, in Cohee, G.V Glaessner, M.F., and Hedberg, H.E. (eds.) The Geologic Time Scale. American Association of Petroleum Geologists Studies in Geology 6. Hazel, J.E., Edwards, L.E., and Bybel, L.M. (1984). Significant unconformities and the hiatuses represented by them in the Paleogene of the Atlantic and Gulf Coastal Plain province. American Association of Petroleum Geologists Memoir 34, 59-66. Holmes, C.W., and Martin, E.A. (1977). Rates of sedimentation, in Berryhill, H .L. (ed.), Environmental Studies, South Texas Outer Continental Shelf, 1976-Geology: Report to the Bureau of Land Management. Horne R.A. (1969). Marine Chemistry: The Structure of Water and the Chemistry of the Hydrosphere. New York: Wiley-Interscience. Hower, J (1961). Some factors concerning the nature and origin of glauconite. The American Mineralogist, 46, 313-334. Huddleston, P.F. (1965). Correlation of Upper Eocene and Lower Oligocene strata between the Sepulga, Conecuh, and Choctawhatchee Rivers of southern Alabama. Unpublished master's Thesis Florida Sate University.


Ireland, B.J., Curtis, C.D., and Whiteman, J.A. (1983). Compositional variation within some glauconites and illites and implications for their stability and origins. Sedimentology, 30, 769-786 72 Koch, C.F. (1987). Prediction of sample size effects on the measured temporal and geographic distribution patterns of species. Paleobiology, 13, 100-107. Kornicker, L.S., and Odum, H.T. (1967). Characterization of modern and ancient environments by species diversity in Transactions of the 17th Annual Meeting of The Gulf Coast Association of Geological Sciences, San Antonio, TX. Krumbein, W.C, and Pettijohn, F.J. (1938). Manual of Sedimentary Petrology New York: Appleton-Century-Crofts, Inc. MacNeil, F.S. (1946). The Tertiary formations of Alabama in Southeastern Geological Society Fourth Field Trip Guidebook, Southeastern Alabama. Magurran, A.E. (1988). Ecological Diversity and it's Measurement. Princeton, NJ: Princeton University Press. Mallinson, D.J. (1988) Distribution and Petrology of Glauconitic Sediments in the Miocene Pungo River Formation, Onslow Bay North Carolina Continental Shelf. Unpublished master's thesis East Carolina University. Mancini E.A., and Tew, B.H. (1988) Paleogene Stratigraphy and Biostratigraphy of Southern Alabama: Field Trip Guidebook for the GCAGS-GCS/SEPM 38th Annual Convention. Mancini, E.A., and Tew, B H (1989). Regional Lithostratigraphy and Biostratigraphy in Upper Cretaceous and Tertiary of West-Central Alabama: A Guidebook for the 26th Annual field Trip of the Alabama Geological Society. Mancini, E.A., and Tew, B.H. (1990) Tertiary Sequence Stratigraphy and Biostratigraphy of Southwestern Alabama, A Guidebook for Field Trip 1. 39th Annual Meeting, Southeastern Section, Geological Society of America. Mancini, E.A., and Tew, B.H. (1991). Relationships of Paleogene stage and planktonic foraminiferal zone boundaries to lithostratigraphic and allostratigraphic contacts in the eastern Gulf Coastal Plain. Journal of Foraminiferal Research 21, 48-66.


73 May R.M. (1975). Patterns of species abundance and diver sity in Ecology and Evolution of Communities Cody, M.L., and Thistle, A.B., eds. Cambridge, MA: Harvard University Press. Odin, G.S., and Letolle R. (1980). Glauconitization and phosphatization environments: A tentative comparison in Marine Phosphorite s -Geochemistry, Occurrence, Genesis, Bentor, Y.K., ed Society of Economic Paleontologists and Mineralogists Special Publication 29. Odin, G.S., and Matter, A (1981). De glauconiarum origine. Sedimentology 28, 611-641. Owens, J.P., and Sohl, N.F. (1973). Glauconites from New J ersey-Maryland Coastal Plain: Their K-Ar ages and application in stratigraphic studies. Geological Society of America Bulletin, 84, 28112838. Palmer K.V.W. (1937). The Claibornian Scaphopoda, Gastropoda, and dibranchiate Cepha l opoda of the southern United States. Bulletins of American Paleontology,] pt. 1, 548p., pt. 2, 90 pls. Palmer, K.V.W. and Brann D .C. (1965, 1966). Catalogue of the Paleocene and Eocene Mollusca of the southern and eastern United States, Parts I and II Bulletins of American Paleontology, 48, 1027p Pettijohn F.J. (1957). Sedimentary Rocks. New York: Harper Brothers. Pinson W.H., Jr. (1960). Reliability of gla uconite for age measurement by K-Ar and Rb-Sr methods Bulletin of The American Association of P etroleum Geologists, 44, 1 793-1808. Porrenga, D.H. (1967). Glauconite and chamosite as depth indicators in the marine environment. M arine Geology, _5, 495-501. Rainwater, E .H (1968). Geological hi story and oil and gas potential of the Central Gulf Coast. Trans ac tions-Gulf Coast Association of Geological Societies, 18, 124-135. Rhoads, D .C. (1974) Organism-sediment relations on the muddy sea floor. Oceanographic Marine Biology Annual Review 12, 263-300. Siesser W.G., Fitzgerald B.G., and Kronman D.J. (1985). Correlation of Gulf Coast provincial P a leo gene stages with European standard stages. Geological Society of America Bulletin 96, 827-831.


Simpson, G.G. (1944) Tempo and Mode in Evolution. New York: Columbia University Press 74 Slobodkin, L.B., and Sanders, H.L. (1969). On the contribution of environmental predictability to species diversity in Woodwell G.M. and Smith, H.H. (eds), Diversity and Stability in Ecological Systems, Upton, New York: Brookhaven National Laboratory. Staff, G.M., and Powell E.N. (1990). Local variability of taphonomic attributes in a paraautochthonous assemb lage: Can taphonomic signature distinguish a heterogeneous environment? Journal of Paleontology 64 648 658. Staff, G.M. R J. Stanton Jr., E.N. Powell and H. Cummins (1988) Time averaging, taphonomy and their impact on paleocommunity reconstruction : Death assemblages in Texas bays. Geological Society of America Bulletins, 97, 428-443 Swindel, D.B (1986). A paleoecological study of the Gosport Sand (Claibornian: Middle Eocene) in Southwest Alabama. Unpublished master's Thesis University of Alabama. Tanner, W.F. (1982). Sedimentological tools for id entifying depositional environments in Arden, D.D. Beck, B.F. and Morrow, E., (eds.), Proceedings: Second Symposium o n the Geology of the Southeastern Coasta l Plain. Toulmin, L.D., LaMoreaux, P E., and Lanphere C.R. (1951). Geology and g round-water resources of Choctaw County, Alabama Geological Survey of Alabama. Special Repo rt 21 and County Report 2. Toulmin, L.D. and LaMoreaux P.E. (1963). Stratigraphy a lon g the Chattahoochee River, connecting link between Atlantic and the Gulf Coastal Pl a ins Bulletin of the American Association of Petroleum Geologists, 47, 385-404. Toulmin, L.D. (1977). Stratigraphic distribution of Paleocene and Eocene fossils in the eastern Gulf Coast region, Geological Survey of Alabama Monograph 13, 1-2. Vernon, R.O. (1951). Geology of Citrus and Levy Counties, Florida Geological Survey Bulle tin 33.




76 APPENDIX 1. LITHOLOGIC DESCRIPTIONS OF STUDIED OUTCROPS Outcrop #1 A road cut 0.4 miles east of Puss Cuss Creek on County Road 14, between Melvin and Gi l bertown, Choctaw County, Alabama in theSE 1/4 of Sec. 28 T11NR4W. Formation Thi ckness Lithology (e m ) Moody's 23 Lime sto ne s a nd y packstone te x ture glauconitic pale yellow i s h orange Bra n c h (lOYR 8/6), forming a hard ledge Gosport 4 0 Sand quartz 75 % subrounded with cla y limoniti c weathered 1 5 % San d dark yellowish orange (lOYR 6/6) to moderate reddish brown (lO R 4 / 6) and s hell ha sh m oll u sca n and ec hin oid, 10 % Burr owed with glauconitic sa ndy lime s tone from above ex tending into burrows. Lower co ntact n o t exp osed.


77 APPENDIX 1 (Continued) Outcrop #2 A road cut on the north side of U.S. Highway 84, 4.3 miles west of Silas Choctaw County, Alabama Fo rm a t io n Thi c kn ess L i t h o l ogy ( em ) Mood y's 4 5 Lim esto n e, san d y pac ks to n e t exture g l a u co nit ic pale yellowish ora n ge ( l OYR B ranc h 8 / 6) fo rmin g a hard l e d g e Co nt ains a b undan t ec hin oids ( Peria r c hus l yelli) Gosport 1 6 5 Sa nd q ua rtz, 75%, s u b r o und e d with clay limo nitic, weathe r ed 1 5%, dark Sa nd yellowis h o r a n ge (lOYR 6/6) to mod e rate r e d d i sh b r own ( l O R 4 / 6) and s h ell h as h m o llu sca n a nd e c hin oid, 1 0 % Burrowed wit h glauco n it i c sa nd y limest o n e from a b ove ex t e ndin g int o b urr ows. Low e r c o nta c t n o t ex'Posed.


78 APPENDIX 1 (Con t inued) Out cro p #3 Gop her o r Bakers Hill, 0.75 mi l es above St. S t ephen's quarry on t h e Tombi g bee R iver, W ashing ton County Alabama in TINR l W at the south end of the river cut-off. Formation Thickn ess Lithology (em) Mood y's 35 Lim esto ne, sa nd y p ackstone text ur e, g l a u co n it ic, pale yellowish ora nge (10YR Bran c h 8/6), forming a h ar d l edge. Co n tains abundant echinoids ( P eriarchus lyelli ). Gosp ort 95 Sand, quartz 65%, grayish orange ( lOYR 7/4) with shell has h a nd echinoids Sand ( P e riar chus lyelli) 20%, a nd lim esto n e stringe rs, 1 5 %, sa nd y p acks t o n e, pa l e yellowish orange ( 10YR 8/6), 2-4 e m thick interbedded 85 Sa nd quartz 75% glauconitic dark yellowish orange (lOYR 6/6), wit h s h ells and shell f ragments, m o llu sca n 25%, and nu merous poorly preseiVed b u rrows. 20 Sands t one m oderate yellowis h brown, (10YR 5 / 4), calcareous foss i l iferous f o rms a hard le dge. 135 Sand, quartz 65%, glauconit ic, dark yellowish o range (lOY R 6/6), with shells and shell fr agme nt s m o llu scan 35%, well preseiVed 50 Sand quartz, 90%, very g l auconitic m oderate yellowish brown ( l OYR 5/4), with s hell fragme nt s m ollu sca n 1 0 %, abra d ed. Well p r eseiVed burrows (Ophi omorp h a) with very g lau conitic walls. 140 Sand, quartz, 75%, g l aucon iti c d ark yellowish o range (lOYR 6/6), and shell fragment s, mollu scan, 25%, abraded and degraded by diss o luti o n including a bundant Venericar dia alticostata, poorl y pres eiVed. Lisb o n 35 Clay, 85%, medium bluish gray (5B 5/1), stiff h a rd wit h sand, quartz, 15%, shelly a nd glauconitic, as above, in b urr ows from above.


79 APPENDIX 1 (Continued) Outcrop #4 Along Little Stave Creek, a westward flowing tributary of Stave Creek 3 5 miles north of Jackson and west of U .S. Highway 43, C l arke County, in Sec. 19 and 20, T7NR2E. Formation Thickness Litho l ogy (em) M oody's 25 Lim estone, sandy, packstone texture glauconitic pale ye llowi sh orange (10Y R Branch 8/6), forming a hard co ncretionary led ge. Contains abundant echinoids (Periarchus lyelli) Gosport 105 Sand quartz, moderate reddish brown (lOR 4/6), with shell hash sparse. Sand 50 Sand, quartz, 85%, glauconitic dark yellowish o range (lOYR 6/6), with shells and shell fragments, 15 %, molluscan and numerous poorly preseJVed burrows. 25 Sandstone, moderate yellowish brown, (lOYR 5/4), cal careo u s, f ossilifero us forms a hard ledge 50 Sand quartz, 55%, gla u conitic lig h t brown (5YR 5/6), with s hell s and shell fragments molluscan 45 %, we ll preseJVed den sely packed. 110 Sand, quartz, 65%, glauconitic, medium dark gray (N5) with s hells, mollu scan, 35%, well preseJVed including abundant Venericardia a l ticostata. 20 Sand, quartz, 55%, phosphatic silty, medium dark gray (N5) with shells and shell fragments, molluscan 45 %, and numerous shark and r ay teeth. Lisbon 35 Clay 95%, o l ive gray (5Y 4/1), hard blocky with sand quartz, 5%, shelly and glauconitic as above, in burrows from above.


8 0 APPENDIX 1 ( C ontinued ) O u t c r o p #5 Gosport Landing at a h i gh b l uff on the right b a nk of the Al a b ama River 4 5 miles bel ow the bridge on U.S. H i ghwa y 8 4 C l a rke C o unty i n the NW 1 / 4 of Sec. 28, D N R5E Formation Thic kn ess Lith o l ogy (em ) Moody's 55 Lime s tone sandy packsto n e texture glauco ni tic pa l e yellowis h orange (10 Y R Branc h 8/6), f orming a hard led ge. Conta in s abund a n t ec hin o id s ( P eria r c h u s l yelli). Gosport 250 Sa n d q u artz, 65%, dark yellowis h oran ge ( lOYR 6/6) wit h s h ell has h a n d Sand echinoids ( P eriarchus lyelli), 35%. 350 San d q u artz 75%, gla u co nitic, dark yellowis h orange (lOYR 6/6), with s h ells a n d s h ell fragme n ts mo llu sca n well p r eserved and d ense l y pac k ed. 15 Sa n dst o ne, moderate yellowis h b r own, (lOYR 5/4), calcareous, fossilifero u s, forms a hard l edge. 35 Sa nd q uartz, 65%, g l a u conit ic, dark yellowis h orange (lOYR 6/6), wit h s h ells and s h ell f ragm e nts, moll u scan 35%, well preserved 350 Sa n d quartz, 80%, very gla u conit i c moderate yellowis h b rown (lO Y R 5 / 4) with s h ell fragments m ollusca n 20 %. Well preserved b u rrows (Oph io morpha) wit h very glauco n itic walls. 190 Sa n d, quartz, 75%, g l a u conit i c d a r k yellowis h orange (10YR 6/6), wit h s h ells a n d shell fragme nt s m ollu sca n 25%, includi n g abunda n t Ven erica r d i a a lti costata well preserved. 115 Sand quartz 95%, s ilty, dar k yellowis h o r ange wit h s h ell h ash 5 % 35 Sa n d, lig ht olive gray (5 Y 5/2), s l ightl y foss i l iferous, l owe r contac t burrowed. Lisbon 45 Clay, 95% greenis h g r ay (50 6/1), stiff, hard with sand q uartz, 5%, s ilt y as above in b u rrows fro m above.


81 APP ENDIX 1 (Continued) Outcrop #6 Claiborne Bluff at a high bluff on the l eft bank of the Alabama River 0.25 mile below the bridge on U .S. Highway 84, Monroe Cou nty, in the NE 1/4 of Sec 30, TINR6E. Formation Thi ckness Lit h o l ogy (e m ) Mood y's 30 L i m esto ne, sa ndy, packsto n e t extu r e glauco n itic p ale yellowis h o r a n ge (10YR B ra n c h 8/6), form ing a h ard l e dge. Co ntains abundant e c h inoids ( P eria r chus lyelli ) G osport 6 5 0 San d qu artz, 55% dark yellowis h ora n ge ( 10YR 6/6) glau conit ic, with s h ell a nd Sa n d s hell fragme nt s 45%, densel y p acke d slight l y carb o n aceo u s ( l ea v es and peat la yers < 1 em thi ck). 55 S and qu a r tz 65%, m edi um bluis h gray (5B 5/1), clayey, with shells mollu sca n 35%, d e n se l y packe d, lower cont act burrowed. L isbon 35 C l ay, 85%, m e dium bluish g r ay (5B 5/1), s tiff hard wit h sand qu a rtz 15 % shelly and glau conitic, as abov e, in burrows fro m abo v e


82 APPENDIX 1 (Continued) Outcrop #7 R attlesnake B luff on the lef t bank of the Alabama R i v er 6 5 miles below the bridge on U .S. Highw a y 84, Monroe C ounty, i n theSE 1 / 4 of Sec 31, TINR5 E. Formation Thick n ess Lit h o l ogy (e m) Moody's 25 L i mes t one, s andy p ac kst o n e t ex tu re, gla u coniti c pa l e y ellowish oran g e ( l O Y R Bra n c h 8 / 6) f orming a h a rd l e d ge Gosport 2 1 0 Sand, q u artz 7 5%, d a rk y e llowis h o r ange (lOY R 6 / 6) w ith s h ells a n d s h ell Sand fragm e n t s poor l y p r eserve d 25%. 50 S a n d q uart z 85%, g l a u coniti c, cla y ey grayis h blue (5 P B 5/2 ) wit h s h ells and s h ell fragments moll u scan an d numero u s s hark teeth, 25%, d e n s e l y pac k e d L i s b o n 1 5 C l ay 85%, m e dium b luis h gray (5 8 5/1), stiff hard with s and, q u a r tz, 1 5 %, s h elly an d g l a u c onit i c as a b ove in b u r r ows fro m abo v e.


APPENDIX 2. OCCURRENCE OF MOLLUSCAN FOSSILS IN CLAIBORNE BLUFF SAMPLES SAMPLE SAMPLE SPECIES* SIZE (I) NUMBER 1 2 3 4 5 6 7 8 9 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** 5 2 CL-6 237 129 351 15 174 71 77 116 6.5 CL-7 94 397 216 11 33 53 6.8 CL-8 133 102 312 26 181 101 80 62 6.5 CL-9 266 124 261 27 136 40 130 111 4.5 CL-10 27 76 177 9 35 46 59 32 5.6 CL-11 197 184 168 16 376 38 272 34 5.0 CL-12 169 37 101 6 252 5 57 33 2.0 CL-13 67 32 20 110 95 10 63 10.5 CL-14 11 91 71 47 17 264 3.9 CL-15 32 7 57 13 20 122 86 3 8 CL-16 42 201 11 38 63 3.2 CL-17 47 17 29 1 56 16 20 9 2.3 CL-18 74 47 26 39 4.5 CL-19 186 11 321 12 68 32 43 4.0 CL-20 57 79 240 70 21 16 4.5 CL-21 3 51 107 24 34 5 13 5.0 CL-22 25 56 12 31 9 2 5 0 CL-23 19 39 18 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** 2 3 4 5 6 7 8 9 *SPECIES (ranked by total abundance in all samples) 1 Turritella carinata !.Lea 2 Venericardia (Pieuromeris) parva Lea 3 Gratelupia (CytheriQpsis) hydana (Conrad) 4 Glycymeris trigonella (Conrad) 5 Mesalia vetusta (Conrad) 6 Linga (Cavilinga) pomilia (Conrad) 7 Glycymeris trigonella minor (Lea) 8 Venericardia (Venericardia) rotunda Lea g Callista (Costacallista) aequorea (Conrad) 83


84 APPENDIX 2 (Continued) SAM P LE SPECIES* NUMBER 1 0 11 1 2 13 14 15 16 17 18 19 20 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** CL-6 1 1 7 128 1 92 6 188 21 1 39 CL-7 13 28 41 18 9 CL8 75 54 43 25 31 18 35 C L -9 58 86 85 15 43 30 68 CL-10 30 23 57 2 54 34 30 CL-11 68 97 38 8 CL-12 27 39 5 6 CL-13 1 9 144 19 19 74 28 15 CL-14 6 66 101 93 8 14 C L15 43 2 32 38 7 28 4 CL-16 11 16 23 CL-17 1 0 5 7 19 3 C L -18 25 36 32 28 14 38 167 C L -19 43 42 60 46 13 43 13 CL-20 15 44 35 24 3 45 13 CL-21 7 26 18 8 13 33 4 CL-22 8 4 2 20 C L -23 5 5 ****** ***** ***** ***** ***** ***** **** ***** ***** ***** ***** ***** ***** 1 0 11 12 13 14 15 16 17 18 19 20 *SPE C I ES 1 0 Retusa (Cylichnina) galba (Conrad) 11 Callista (Callista) perovata (Conrad) 1 2 Caryoco r bula alabamiensis g r egorioi (Gassmann) 13 Spisu l a pa r i l is (Co n rad) 14 Caestocorbula murchisonii (Lea) 15 Diplodonta (Diplodonta) ungulina (Conrad) 16 Tellina (Eurytelli n a) papyria Conrad 1 7 Venericardia (Ciaibornicardia) a l ticostata (Conrad) 1 8 Caryocorbula a l abamiensis ( L ea) 19 Glycymeris staminea (Conrad) 20 Platytrochus stokesi (Lea)


85 APPENDIX 2 (Continued) SAMPLE SPECI ES* NUMBER 21 22 23 24 25 26 27 28 29 30 3 1 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** CL-6 30 13 10 3 64 6 CL-7 CL-8 20 48 7 28 4 14 23 6 9 CL-9 46 31 3 34 24 54 7 CL-10 12 26 10 38 4 30 8 CL-11 28 24 30 2 18 26 CL-12 39 11 6 7 31 CL-13 1 46 2 7 C L14 26 53 CL-15 16 103 3 138 16 CL-16 3 5 3 CL-17 11 8 5 7 CL-18 9 103 CL-19 6 8 3 CL 20 3 15 CL-21 5 7 2 60 CL-22 9 5 37 CL -23 9 2 22 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** 21 22 23 24 25 26 27 28 29 30 31 *SPECIES 21 Linga (Cavi linga) pomilia alveata (Conrad) 22 Crepidula lirata Conrad 23 Egerella limatula (Conrad) 24 Nuculana coelata (Conrad) 25 Nucula (Nucula) magnifica Conrad 26 Parm ico rbula gibbosa (Lea) 27 Turritella obruta Conrad 28 Nucula (Nucula) ovula Lea 29 Bullata semen (I. Lea) 30 Glycymeris idonea (Conrad) 31 Calyptraphorus velatus (Conrad)


86 APPENDIX 2 (Continued) SAM PLE SPEC IES* NUMBER 32 33 34 35 36 37 38 39 40 41 42 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** CL-6 21 8 32 27 1 9 15 13 9 31 CL-7 CL-8 2 18 13 34 24 21 2 5 4 15 CL 9 7 51 32 12 27 15 19 1 3 13 CL-10 5 9 6 6 13 3 14 1 6 CL-11 9 9 2 2 3 2 CL-12 3 10 1 4 5 0 CL-13 18 4 8 5 CL-14 1 69 10 32 CL-15 4 8 3 CL-16 7 2 8 5 CL-17 2 4 2 CL-18 15 14 CL-19 17 7 17 9 CL-20 29 CL-21 8 3 3 3 2 CL-22 4 CL-23 2 3 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** 32 33 34 35 36 37 38 39 40 41 42 *SPECIES 32 Neverita limula (Conrad) 33 Pitar (Pitar) nuttali (Conrad) 34 Agaronia alabamensis (Conrad) 35 Callista (Callista) perovata lisbonensis (Harris) 36 Callista (Callista) perovata subvitrea (de Gregorio) 37 Oentalium minutistriatum Gabb 38 Bathytormus prote xus (Conrad) 39 Agaronia bombylis (Conrad) 40 Plicatula filamentosa Conrad 41 Lirodiscus tellinoides (Conrad) 42 Lucina sp.


87 APPENDIX 2 (Continued) SAMPLE S PECIES* NUMBER 43 44 45 46 47 48 49 50 51 52 5 3 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** C L6 13 28 4 3 30 22 13 5 CL-7 2 CL-8 1 9 11 4 5 2 13 6 12 CL-9 4 18 15 10 5 4 9 5 15 15 CL-10 16 13 5 5 2 4 CL-11 1 13 6 9 3 CL -12 7 1 4 5 2 CL-13 2 7 5 CL-14 3 CL-15 7 8 CL-16 2 1 3 5 CL-17 6 CL-18 CL1 9 5 33 6 2 CL-20 C L -2 1 CL-22 CL23 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** 43 44 45 46 47 48 49 50 51 52 53 *SPECIES 43 Venericardia (Vener i cor) claiboplata Gardner and Bowle s 44 Hinds iella faba donacia Oall 45 "Marginella" constricta Conrad 46 Ancilla staminea (Conrad) 47 "Astarte proruta" Co nrad 48 Limopsis sp. 1 49 Crepidula dumosa Conrad 50 Calorhadia (Calorhadia) opulenta (Con r ad) 51 Penion bellus (Conrad) 52 Buccitriton sagenum (Conrad) 53 "Natica" "(Naticarius)" s emilunata !.Lea

PAGE 100

. APPENDIX 2 (Continued) SAMPLE NUMBER SPECIES* 54 55 56 57 58 59 60 61 62 63 64 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** CL-6 CL-7 CL-8 CL-9 CL-10 CL-11 23 CL-12 6 CL-13 CL-14 CL-15 CL-16 CL-17 2 CL-18 CL-19 CL-20 CL-21 CL-22 CL-23 19 6 2 4 3 16 24 2 4 3 11 2 2 9 5 1 3 3 2 2 3 5 5 2 16 7 5 1 10 13 6 5 4 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** 54 55 56 57 58 59 60 61 62 63 64 *SPECIES 54 Caryocorbula deusseni (Gardner) 55 Chlamys deshayesii (Lea) 56 Uromitra gracilis (H C.Lea) 57 Epilucina rotunda (Lea) 58 Bullata larvata (Conrad) 59 Dentalium blandum deGregorio 60 Callista (Costacallista) aldrichi (Harris) 61 Ar chitectonica fungina (Conrad) 62 Turritella dutexata Harris 63 Sinum declive (Conrad) 64 Athleta sayanus (Conrad 88

PAGE 101

. APPENDIX 2 (Continued) SAMPLE NUMBER SPECIES* 65 66 67 68 69 70 71 72 73 74 75 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** CL-6 C L-7 CL-8 CL-9 C L -10 CL-11 CL-12 CL-13 CL-14 CL1 5 CL-16 CL17 CL18 CL-19 C L 20 C L-21 C L -22 CL 23 5 2 4 4 3 3 4 1 6 4 1 3 4 2 5 2 8 4 6 2 6 2 1 3 4 4 1 2 7 ****** ***** ****** ***** ***** ***** ***** **** ***** ***** ***** ***** ***** 65 66 67 68 69 70 7 1 72 73 74 75 *SPECIES 65 P o lin ices e minulu s (Conrad) 66 Limopsis sp 2 67 Athl e t a petrosus (Conrad) 68 Solariella sta l agmium (Conrad) 69 Lucina (Recurvella) dolabra Conrad 70 Balcis claibornia (Pa lm er) 7 1 T enagodus vitis (Conrad) 72 Car icella bol aris (Conrad) 73 Car ice lla pyruloides (Conrad) 74 T r igonostoma gemma tum (Co nr ad) 75 D oliocassis nupe r a (Co nr ad) 89

PAGE 102

. APPENDIX 2 (Continued) SAMPLE NUMBER SPECIES* 76 77 78 79 80 81 82 83 84 85 86 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** CL-6 CL-7 CL-8 CL-9 CL-10 CL-11 CL-12 CL-13 CL-14 CL-15 CL-16 CL-17 CL-18 C L-19 CL-20 CL-21 CL-22 CL-23 4 1 2 2 5 2 3 1 6 2 1 2 4 3 3 1 5 3 2 5 2 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** 7 6 77 78 79 80 81 82 83 84 85 86 *SPECIES 76 Cryptospira (Euryentome) silabra (Palmer) 77 Serpulorbis squamosus (Conrad) 78 Cyclostremiscus exacuus (Conrad) 79 Cerithiella nassula (Conrad) 80 Pseudoliva vetusta (Conrad) 81 Turbindia pharetra Lea 82 Diodora tenebrosa (Conrad) 83 Serpulorbis major (Chavan in Palmer) 84 Architectonica (Granoso lar ium) ornata (I.Lea) 85 Barbatia (Piagiarca) rhomboidella (Lea) 86 Ranellina maclurii (Conrad) 9 0

PAGE 103

APPENDIX 2 (Continued) SAMPLE NUMBER SPECIE S 87 88 89 90 91 92 93 94 95 96 97 ****** ***** ****** ***** ***** ***** ***** *** ** ***** ***** ***** ***** ***** C L-6 C L -7 C L-8 C L9 CL-10 CL-11 CL-12 C L-13 CL-1 4 CL-15 CL-16 CL-17 CL 18 CL1 9 C L-20 C L2 1 CL-22 C L -23 2 1 2 2 2 3 2 3 2 2 2 ****** ***** ****** ***** ***** ***** **** * **** ***** ***** ***** ** ** ***** 87 88 89 90 9 1 92 93 94 95 96 97 *SPECIES 87 Architectonica (Solar i axis) elabo rata (Co nra d) 88 D entalium thalloides Conra d 89 M azza lin a i n au r ata (Conrad) 90 Pte r ia limula (Con r ad) 9 1 T e n agod u s vitis (Conrad) 92 Pen io n b ellus leai (Palmer) 93 L evifus u s mortonii ( I.Lea) 94 Conom itr a fusoides ( I.Lea) 95 Eopleurotoma sayi (I.L ea) 96 So l a r iella tr i costata (Conrad) 97 Levifusus trabeatus (Conrad) 91

PAGE 104

APPENDIX 2 (Continued) SAMPLE NUMBER SPECIES* 98 99 1 00 1 01 1 02 1 03 1 04 1 05 1 06 1 07 1 08 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** CL-6 CL-7 CL-8 CL-9 CL 10 CL-11 CL-12 CL-13 CL-14 CL-15 CL-16 CL-17 CL-18 CL-19 CL-20 CL-21 2 CL-22 CL-23 ****** ***** ****** ***** ***** ***** ***** ***** ***** ***** ***** ***** ***** 98 99 1 00 1 01 1 02 1 03 1 04 1 05 1 06 1 07 1 08 *SPECIES 98 Turritella nasuta Gabb 99 "Hastula" venusta (l.lea) 100 Conomitra fusoides I epa de Gregor io 1 01 Norrisia (Norrisella) micromphala (Gassmann) 1 02 Latirus extricatus (Casey) 1 03 Cornulina sp. 104 Chlamys sp. 1 05 Solariella cancellata (Conrad) 1 06 Architectonica (Stellaxis) alveata (Conrad) 1 07 Serpulorbis squamulosus (Conrad) 108 Lirodiscus sp. 92

PAGE 105

93 APPENDIX 2 (Continued) SAMPLE SPECIES* NUMBER 1 09 110 111 112 ****** ***** ****** ***** ***** ********* CL-6 CL-7 CL-8 CL-9 CL-10 CL-11 CL-12 CL-13 CL-14 CL-15 CL-16 CL-17 CL-18 CL-19 CL-20 CL -2 1 CL-22 CL-23 ****** ***** ****** ***** ***** ********* 109 110 111 112 *SPECIES 109 Lacinia alveata (Conrad) 11 0 Turridae sp 111 Hipponix pygmaeus I.Lea 112 Eomiltha pandata (Conrad)


Download Options

No images are available for this item.
Cite this item close


Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.


Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.


Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.


Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.