0 X.' 1. ~ 494'*X' 4~ ·t _l of the FLORIDA STATE MUSEUM Biological Sciences Volume 19 1975 Number 2 DIVERSITY AND COMMUNITY STRUCTURE OF THE BRUSH CREEK MARINE INTERVAL (CONEMAUGH GROUP, UPPER PENNSYLVANIAN), IN THE APPALACHIAN BASIN OF WESTERN PENNSYLVANIA GRAIG D. SHAAK 2- r - * bA' f . 4..1.*LY' f~ 2.1 4 .r , . - 7» 3 f Ij , f i:-S :; J. j . 1 I - UNIVERSITY OF FLORIDA GAINESVILLE Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCIENCES, are published at irregular intervals. Volumes con- tain about 300 pages and are not necessarily completed in any one cal- endar year. CARTER R. GILBERT, Editor RHODA J. RYBAK, Managing Editor Consultants for this issue: GARRY G. ANDERSON DAVID NICOL Communications concerning purchase or exchange of the publications and all manuscripts should be addressed to the Managing Editor of the Bulletin, Florida State Museum, Museum· Road, University of Florida, Gainesville 32611. This public document was promulgated at an annual cost of $2,249.10 or $2.249 per copy. It makes available to libraries, scholars, and all interested persons the results of researches in the natural sciences, emphasizing the Circum-Caribbean region. Publication date: January 6, 1975 Price: $2.30 DIVERSITY AND COMMUNITY STRUCTURE OF THE BRUSH CREEK MARINE INTERVAL ( CONEMAUGH GROUP, UPPER PENNSYLVANIAN), IN THE APPALACHIAN BASIN OF WESTERN PENNSYLVANIA GRAIG D. SHAAKl SyNopsIs:. Nine Brush Creek localities ih western Pennsylvania were trench sampled for the purpose of recovering the total fauna from each visible marine horizon. The 9 localities provided 61 samples that were disaggregated by the Amine 220 tech- nique, washed, dried, and sieved into 305 individual samples. The recovered fossils are judged to have lived, died, and been preserved at each collecting site and are therefore members of fossil communities. This judgement is based on qualitative judgements of state of articulation, abrasion level, quality of preservation, and taxo- nomic associations. Animals from the sieve-sample splits were viewed microscopically to determine taxonomic composition and frequencies. Raw faunal assemblage data were adjusted to reduce the frequencies to whole numbers of individuals and, secondly, to a single generation. The purpose of making a generation correction is to reflect standing crop, which in turn provides a reasonable approximation of the once-living communi- ties. The adjusted outcrop data were examined for faunal similarities among localities for the purpose of zoning the localities into workable groups. Two distinct faunal groups are recognized on the basis of relative water depths and distances from shore. Four shallow water and five deep water localities are inferred. The deep-water localities are characterized by low diversity molluscan faunas, whereas the shallow- water localities are characterized by high-diversity molluscan faunas. The adjusted data were further subjected to programmed diversity and equitability tests to decipher faunal succession and community structure. The diversity and equitability data were graphed for each locality and the resultant curves compared with ecological criteria such as animal habitat preferences and feeding types. These combined types of data aided in deciphering the sea-level history of the marine event at each locality. Transgressive, stillstand, and regressive phases are deciphered, along with their corresponding opportunistic, stable-mature, and relict-mature faunas. These data also show that the faunal distribution above and below the stillstand is asymmetrical. i The author is Assistant Curator in Invertebrate Paleontology, Florida State Museum, and Assistant Professor in the Department of Geology, University of Florida, Gaines- ville 32611. Manuscript accepted 25 February 1974. SHAAK, GRAIG D. Diversity and Community Structure of the Brush Creek Marine Interval ( Conemaugh Group, Upper Pennsylvanian), in the Appalachian Basin of Western Pennsylvania..Bull. Florida State Museum, Biol. Sci., Vol. 19, No. 2. pp.69- 133. 70 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 TABLE OF CONTENTS IN'rRODUCTION __- 70 Geographic and Stratigraphic Setting 71 Previous Investigations 74 ACKNOWLEDGMENTS 75 METHODS OF INVESTICATION 75 Field Sampling --- 75 Laboratory Preparation 76 Faunal Analysis _ 76 Data Adjustments to Reflect Standing Crop 77 Diversity Indices, Species Diversity, and Equitability 79 THEORETICAL FRAMEWORK --- 81 Characteristics of Level-Bottom Marine Communities 81 Characteristics of the Brush Creek Depositional Basin 84 FAUNAL SUCCESSION AND DIVERSITY 84 High Diversity Moliuscan Faunas 87 Locality 1, Kittanning, Pennsylvania 87 Locality 7 ( Piccolomini Quarry), Uniontown, Pennsylvania 89 Locality 8, Ufsina, Pennsylvania 91 Locality 9, Glade City, Pennsylvania 93 Low Diversity Molluscan Faunas 97 Locality 2, Shelocta, Pennsylvania 97 Locality 3, Barton, Pennsylvania 99 Locality 4, Sewickley, Pennsylvania 100 Locality 5, Clenshaw, Pennsylvania 103 Locality 6, Murrysville, Pennsylvania 105 The Limestone Problem 108 COMMUNITY STRUCTURE 110 General Statement 110 Astartella-lanthanopsis Community 111 Glabrocingulum-Plagioglypta Community 112 Aclisina Community 114 Rhombopora-Septopora Community 115 DIVERSrry INDICES AS PALEOECOLOGIC TOOLS 115 General Statemeht 115 Comparison of the Shannon-Wiener and Simpson Diversity Functions 116 Restrictions in the Use of Diversity Indices 117 CONCLUSIONS 118 REFERENCES CITED - -119 APPENDIX A. Measured Sections Showing Sampling Interval at Sample Sites _-123 APPENDIX B. Faunal Data Matrix 132 INTRODUCTION The purpose of the investigation was to apply and test modern paleo- ecological and ecological concepts and information theory in attempting to resolve the present problem. It is a segment of a long-range investi- gation, already in progress, by several faculty members of the Depart- ment of Earth and Planetary Sciences, University of Pittsburgh, to reconstruct and interpret the shallow benthic marine communities for all the marine events represented in the Conemaugh Group. Compari- 1975 SHAAK: BRUSH CREEK 71 sons will later be made between the Conemaugh communities and con- temporaneous faunas of similar lithologies in the more western basins of Pennsylvanian age. The problem is one of reconstructing and interpreting the shallow benthic marine communities of the Brush Creek marine interval, Cone- maugh Group ( Upper Pennsylvanian) in the appalachian basin of west- ern Pennsylvania. The Brush Creek marine event represents one of a series of four temporally limited marine events in a succession of prin- cipally non-marine events represented in a cyclothemic sequence, the Conemaugh Group. GEOGRAPHIC AND STRATIGRAPHIC SE'ITING The study area is that part of the Appalachian basin lying within the Pittsburgh · Plateaus section of the Allegheny Plateaus physiographic province. The study area lies between 79°00' and 80°30' west longitude and 39°30' and 41°00' north latitude. It encompasses 3,960 square miles. Nine localities in six western Pennsylvania counties ( Armstrong, Indiana, / · SCALE c 036 12 1. 24 30 36 111 1 1 1 1 + BUTLER -- --4 ARMSTRONG I Measured Sections /1 ? t V A · ; O H IO 03 i1 1 ./ .2 7BEAVER 1-__._____._ 3 „ , ! / INDIANA ,/ i-i I l. 1 . ..7 ~ ,~ CAMBRIA 1 ' - . 31 r ALLEGHENY ,~6 / . :>- / i1 , / 1/ IWESTMORELAND #.- I! 2 1 L----4 WASHINGTON ~--6- j f / t -' C /-< ,40°. 6 +!+-- .1 . I/7 / SOMERSET GREENE ,) / FAYETTE i.8 .9 r 5 I W., VA. MD. , 800 790 FIGURE 1.-Brush Creek locality map, western Pennsylvania. 72 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 Allegheny, Westmoreland, Fayette, and Somerset) were measured and trench sampled in this investigation ( Fig. 1). The Pennsylvanian System was named by Williams ( 1891) for the state of Pennsylvania in the central Appalachians, replacing the out- moded name «Coal Measures." The Pennsylvanian System is divided into the Pottsville, Allegheny, Conemaugh, and Monongahela Groups in ascending order. The Conemaugh Series was named by Platt ( 1879) for the Cone- maugh River valley of western Pennsylvania ( Fig. 2). The Conemaugh ' ,600 800 O H IO 900 r 7 I1 Ji 4: L W. VA. MD. ,·mmm,m,1 }~Conemaugh Group FIGURE 2.-Distribution and isopachous map of the Conemaugh group, western Pennsylvania ( adapted from Branson 1962). includes the strata from the top of the Upper Freeport coal ( Upper Allegheny) to the base of the Pittsburgh Coal (Lower Monongahela). These boundaries correspond with those of the Lower Barren Coal Mea- sures, as previously named by Rogers ( 1858). Flint ( 1965) proposed subdividing the Conemaugh Group into lower ( Glenshaw) and upper ( Casselman) formations for their type sections in the Glenshaw vicinity, Shaler Township, Allegheny County, and along the Casselman River in Somerset County, Pennsylvania, and Garrett County, Maryland, respec- tively ( Fig. 3). This nomenclature is adopted for this study. 1975 SHAAK: BRUSHCREEK 73 Pittsburgh coal bed N O ILV W H O l N V W l jl S S V D N O ILV W W O J M V H S N 3 1 9 Marine beds lacking; freshwater limestones relatively abundant C O N E M A U G H G R O U P 11 Ames limestone (marine) Woods Run limestone (marine) 1 1 Pine Creek limestone (marine) Buffalo sandstone Brush Creek shale 11 marine Brush Creek limestone interval Brush Creek shalt Brush Creek coal Upper Freeport cool bed FIGURE 3.-Subdivision of the Conemaugh group and generalized Brush Creek in- terval ( adapted from Flint 1965). 74 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 The Brush Creek Member is the lowest of four marine units in the Glenshaw Formation and was named by I. C. White ( 1878) for exposures along Brush Creek, Cranberry Township, Butler County, Pennsylvania. The Brush Creek is correlated with the Kansas City Group, Missourian age, of the northern midcontinent area, based on the common occurrences of the fusulinid Triticites ohioensis ( Moore 1944: 684). PREVIOUS INVESTIGATIONS Other than general faunal lists, there is a paucity of published paleon- tological studies of the Conemaugh Group. Mark ( 1912) prepared the first comprehensive taxonomic treatment of Conemaugh fossils from Ohio. Many of her identifications were documented from Raymond's ( 1910) preliminary faunal list of Allegheny and Conemaugh fossils in western Pennsylvania. Burke ( 1930) described the Ames Limestone fauna from Painter Hollow, West Virginia. The Ames fauna of the Pittsburgh Quadrangle was described by Theiss ( 1940). Seaman ( 1940, 1941, 1942) published short papers on the Ames, Pine Creek, and Brush Creek limestones and their associated faunas of western Pennsylvania. Miller and Unklesbay ( 1942) prepared a comprehensive faunal study of cephalopods from Conemaugh rocks of western Pennsylvania. Leigh- ton ( 1947) prepared a guidebook for Pittsburgh geology in which he figured 21 Conemaugh taxa and prepared an extensive faunal list of Conemaugh invertebrates. Doney ( 1954) figured and described 11 Brush Creek bivalve species from the Pittsburgh area and the immediate vicinity. Lintz ( 1958) published on the fauna of the Ames and Brush Creek shales of western Maryland. Murphy ( 1967) published one short paper on the Brush Creek bivalves from Ohio and another ( 1970) on the Brush Creek coiled nautiloid cephalopods from eastern Ohio -and western Pennsylvania. One of the most valuable taxonomic aids used during this study was Pennsyloanian Brachiopods of Ohio ( Sturgeon and Hoare 1968). Hos- kins' ( 1969) Fossil collecting in Pennsylvania and Wagner et al.9 (.1970) Geology of the Pittsburgh area were two very general but valuable taxonomic aids. Paleoecology, although a youthful branch of geology, is becoming an extremely valuable tool in deciphering earth history. Several recent ex- cellent paleoenvironmental studies have been conducted that are of great aid in paleoecological reconstructions. For example, Morris ( 1967) prepared a very detailed and comprehensive environments-of-deposition and paleogeographic study of the Appalachian basin coal fields. Griese- 1975 SHAAK: BRUSH CREEK 75 mer ( 1970) recognized seven faunal assemblages in the Ervine Creek Limestone ( Late Pennsylvanian) of the midcontinent region. He com- bined both lithologic and paleontologic criteria for this reconstruction. One of the more successful areas of paleoecological research has in- volved the Pennsylvanian shallow marine benthic faunas. Johnson's ( 1962) pioneering statistical approach to interspecific associations ( often termed fossil communities ) with Pennsylvanian assemblages is an ex- cellent work, and it is one of the most frequently cited papers among the more recent literature. Fagerstrom ( 1964) described four fossil assemblages from the Pennsylvanian of Nebraska, two of which were named communities ( fossil community and residual fossil community). Stevens ( 1965, 1966, 1969, 1971) recognized communities from fossil assemblages of the Pennsylvanian Minturn Formation in Colorado and the Permian rocks of Nevada and Utah. Other important Pennsylvanian studies are by Shabica ( 1970), West ( 1970), Williams ( 1960), and Zangerl and Richardson ( 1963). Many recent paleoecological endeavors have become quite sophisti- cated in their approach. Cybernetics, stability, diversity, community dynamics, and other concepts in modern ecological theory are being tested in the fossil record. Species diversity and equitability gradienls as paleoecologic tools have been applied to carboniferous studies by Beer- bower and Jordan ( 1969), Donahue and Rollins ( in press), Donahue, Rollins, and Shaak ( 1972), Rollins and Donahue ( 1971, 1972), and West ( 1970). ACKNOWLEDGMENTS I wish to express my graditude to H. B. Rollins, N. K. Flint, J, Donahue, and R. Lund, University of Pittsburgh, and H. Buchanan, University of West Virginia, for their guidance and encouragement during the course of this investigation. I wish to acknowledge V. Schmidt of the University of Pittsburgh for programming the diversity indices. This investigation was in part supported by the Society of the Sigma Xi grants- in-aid of research, National Science Foundation Research Grant GA-31898 to H. B. Rollins and J. Donahue, and the Department of Earth and Planetary Sciences, University of Pittsburgh. METHODS OF INVESTIGATION FIELD SAMPLING The stratigraphic section at each of nine selected field localities was measured and described. Stratigraphic descriptions were recorded for the purpose of defining major lithologic units. The lithologic units complement the faunal analyses in deciphering paleoenvironments in that 76 BULLETIN FLORIDA STATE MUSEUM Vol. 19, N6. 2 the environments of deposition can sometimes be interpreted from the resultant lithologies. The sampling method selected for this study is the stratiBed sampling technique described by Griffiths (1967: 18.). The visible marine zone ( the vertical extent of macrofossils) at each locality was trench sampled from top to bottom. Individual subsamples were taken within single lithologies. In relatively thick homogeneous lithologies one foot intervals were sampled, with approximately five pounds of rock taken in each bulk sample. The entire trenched vertical marine horizon containing macro- fossils was brought back to the laboratory. The reason for this form of detailed sampling is to eliminate the effect of errors or biased samples. In addition to these stratifed samples, several spot samples were taken at each locality to aid in the identification of the fauna. Observations on the state of fossil preservation were recorded. LABORATORY PREPARATION The bulk random samples were brought to the laboratory, and the total samples treated by the Amine 220 technique ( Lund 1970: 578). The Amine 220 technique is superior to the use of either kerosene or Quaternary "0" ( Geigy Chemical) for disaggregation of shales and friable limestones. It is not successful, however, in treating dense lime- stones. Bulk samples were left in the solution for an average of two days and were frequently agitated to keep the fines in suspension. Many samples, especially argillaceous shales and mudstones, were completely disaggregated into fossils and mud. The disaggregate was wet sieved through U.S. Standard Sieve series numbers 2, 10, 40, 100, and 120. Numbers 2 and 10 retain the available macrobenthos, number 40 retains the meiobenthos, and numbers 100 and 120 retain much of the microbenthos ( Wigley and Mcintyre 1964). The disaggregate was thoroughly washed until all traces of the Amine 220 were removed. The washing process must be complete or upon drying the individual grains become cemented into a gelatinous mass. A total of 9 localities was trench sampled, providing 61 stratifled samples and 305 individual sieve fractions. FAUNAL ANALYSIS All sieve fractions were split with a standard sample splitter until a workable quantity was reached. A workable quantity was determined from a preliminary microscopic count of the fossils from the individual sieve fractions. 1975 SHAAK: BRUSH CREEK 77 Samples from each locality were analyzed, one sieve size at a time. The + 40 splits were analyzed first because this size fraction ·contained an admixture of the larger microfossils and smaller macrofossils. The + 40 splits were followed in order by the + 100, + 120, + 10, and + 2 splits. An artificial subdivision of micro-macrofossils, based on size, was established for the convenience of data accumulation and processing. All fossils retained on the number 40, and coarser, sieves were deemed macrofossils,·and all those passing through the number 40 sieve, micro- fossils. A stereoscopic microscope was used for identification and counts of the faunal elements, the microfossils at 4OX and lOOX magnification and the macrofossils at 25.6X. Whenever possible, fossils were identified to the generic level. Owing to their disarticulated state, echinoderms were identified to class. Sponge spicules, conodonts, and vertebrates, all of which were fragmentary, were identified to phylum. Bryozoans, except for two readily recognizable genera, were identified only to growth form. All fragmentary fossils greater than 50 percent complete were counted, and the identification and frequencies were recorded on data sheets for subsequent analyses. The counting of the individual sieve size splits was stopped when a qualitative guess as to adequacy of sample size was reached. This guess was adapted from the rarefaction method arithmetic plots of Sanders ( 1968). If the number of species is plotted against their frequency, the curve initially parallels the ordinate be- cause of the high frequency of new species. As fewer and fewer new species are added, the frequencies of already identified species are still increasing and the curve now asymptotically approaches the abscissa. The inflection point of the curve was used to guess the adequacy of sample size. In most cases an adequately sized sample was reached within the range of the initial sample splits but in several samples the remaining fraction had to be resplit and counted to achieve adequacy of sample size. The equitability of each split was subsequently quanti- tatively determined in conjunction with the diversity indices. DATA ADJUSTMENT TO REFLECT STANDING CROP Ager ( 1963: 184) provides a comment on the tenuity of reconstruct- ing fossil communities from fossil assemblages: 'The Paleoecologist must never forget that he is studying not the living inhabitants of the village but only the bodies in the churchyard, and then only after many visits by grave robbers." The fossil assemblage, as quarried from the outcrop, may reflect little 78 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 about its predecessor, the living community. However, by adjusting the raw faunal assemblage data to reflect standing crop, a realistic approxi- mation of the once living community is possible. One important condi- tion that must be met before making these adjustments is that the as- semblage cannot be biased by transport and mixing. Indicators of trans- portation, such as the right-left value-sorting phenomenon, size sorting, and abrasion were judged at the outcrops and in the washed samples to be at a minimum. The concept of post-mortem transportation has long been a deterrent to paleoecological reconstructions. Johnson ( 1972: 152) stresses this overemphasis and states that transportation is rare. He also provides a most refreshing statement: "It is about time that paleontol- ogists stop apologizing about this possible source of bias as there are enough probable sources to worry about. Recorded taxonomic frequencies must be adjusted to a common stan- dard to enable comparison not only between the different samples within an outcrop but among outcrops. These adjustments are necessary because of the combined fossil and sediment weight differences of the individual sieve sample splits. Initial frequencies were adjusted to a weight stan- dard of 500 grams. For example, if the weight of the counted sample ( both fossils and sediment) was 1.25 grams, the individual frequencies were increased by a factor of 400, which approximated the total that would be found if 500 grams were initially counted. Taxonomic frequencies in raw faunal assemblage data are strongly biased by the various taxa that fragment or disarticulate during taphon- omy. These biases should be compensated for before the data can validly be subjected to statistical tests. The frequencies of fragmented or disarticulated taxa must be reduced to a whole number of individuals. For example, in only the rarest preservations are the radix, column, and calyx of a crinoid preserved in an articulated state. The calyx usually disarticulates into single crystal calcite plates that literally "cleave" themselves out of existence. The column disarticulates into individual, columnals. Inasmuch as these columnals often are the only crinoid remnants in an outcrop, the whole number of individuals must be de- rived from the frequency of individual columnals. Frequencies of cri- noid columnals, sponge spicules, echinoid spines, holothurian sclerites, conodonts, and vertebrate fragments were reduced by a factor of 20 ( Lund, pers. comm.). Bryozoan frequencies were reduced by a factor of 6.3, based on the ratio between the average zoarial lengths of nearly complete specimens on slabs and those retained on the intermediate sized sieve ( + 40). Ostracods average nine ecdysis stages; thus their frequencies were reduced by a factor of nine (Kesling 1961: Q19). 1975 SHAAK: BRUSH CREEK 79 Final adjustments to the taxonomic frequencies are intended to re- duce the values to a "reflection" of standing crop ( Shaak and Rollins 1972). Molluses, which probably enjoy the longest average longevity ( five years), were selected to be the "standard bearer." This five-year average molluscan longevity value is here defined as unity. The fre- quencies of all remaining taxa with either longer or shorter estimated longevity were reduced in proportion of their longevities to the average molluscan value ( Table 1). For example, most foraminifers and ostra- TABLE 1.-SUMMARY OF LONCEVITY VALUES OF BRUSH CREEK TAXA. Taxon Longevity Source Gastropoda small 3 years Comfort 1964: 82-83 large 5 years Comfort 1964: 82-83 Bivalvia epifaunal 5 years Comfort 1964: 84-85 infaunal 3 years Comfort 1964: 84-85 Foraminifera 3 generations Thorson 1971: 152 per year Ostracoda 3 generations Thorson 1971: 152 per year Echinodermata Crinoidea 5 years Fell and Pawson 1966 : 50 Meyer ( pers. comm. ) Echinoidea 5 years Moore 1966: 75 Holothuroidea 5 years Pawson 1966: 63 Asteroides 5 years Feder and Christensen 1966: 109 Ophiuroidea 15 years Fell 1966: 130 Bryozoa 5 yearsl Ryland 1971: 82-85 Boardman and Cheetam 1969: 225 Brachiopods 4 years Hymen 1959: 590 Rudwick 1971: 156 Trilobita 5 years2 C6mfort 1964: 80 Porifera 12 years Comfort 1964: 81 Coelenterata Anthozoa 5 yearsa Wells ( pers. comm.) 1 Based on sexual maturity at five years. 2 Average longevity of Malacostraca. 3 Based on annual growth increments of 20 mm per year. cods average three generations per year; thus their frequencies were reduced by a factor of 15. These longevity data were compiled from estimated values of fossils as well as recent analogs. DIVERSITY INDICES, SPECIES DIVERSITY, AND EQurrABILrry Diversity can vary in two ways; thus there are two different groups of diversity indices ( Donahue and Carothers 1972). One type is the 80 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 numerical percentage composition ( Sanders 1968) of the various taxa in the sample. This is based on numerical dominance. · If all taxa are numerically dissimilar, dominance is high and diversity low. The second variety is kn6wn as species diversity ( Whittaker 1965) and is determined by the number of taxa present in the sample. The more species present in the sample, the higher the diversity. With the two types of diversity and their two types of numerical variance, it was decided to use a repre- sentative diversity index of each type. The Simpson index ( 1949) is calculated from the following equation (N= number of taxa per sample and n = number of individuals per taxon): Ds-N(N-1) In ( n-1) The observed Simpson values range from 1.00 to 6.441. The second index used is the Shannon-Wiener information function (Lloyd and Ghelardi 1964) (s = total number of species and pr = observed proportion of individuals that belong to the r'b species [r= 1,2, . . ., sl): S H(s) --Epr logzpr r-1 The calculated Shannon-Wiener values range from 0.0 to 3.030. Donahue (pers. comm.) developed an equitability test used in this study that not only measures numerical equality or evenness of the Simpson Index distribution but also the adequacy of sample size: where number of taxa the calculated Simpson Index (Ds) for a particular sample is divided by the number of taxa in that sample.' The calculated equitability values range from 1.0 to 0.064. An equitability value of 1.0 denotes a perfect equitable distribution, ,and decreasingly lower values reflect increased dominance. The equitability values also provide a numefical test for adequacy of sample size. The ratio between equitability and the num- ber of taxa in each sample should approximate a value of one, omitting the decimal in the equitability value, if the sample size is adequate. Unfortunately there are no data io define the range of acceptable values in either rejecting or accepting the sample as being of statistically sound size. Because the Simpson index is affected by sample size, a ratio near a value of one suggests that the particular Simpson value used here is not overly biased by sample size. The taxonomic frequency values were plotted for each sample from 1975 SHAAK: BRUSH CREEK 81 each outcrop for total fauna, molluscs only, and without molluscs to aid in the interpretations of diversity and equitability Ructuations. THEORETICAL FRAMEWORK CHARACTERISTICS OF LEVEL-BorroM MARINE COMMUNITIES With respect to hard parts and probability of rapid burial, the marine benthic communities are the most preservable communities of the bio- sphere ( Johnson 1964: 120). If one accepts this premise, the level-bob tom marine community concept should be relatively easy to define, but a review of the literature indicates just the opposite. The term biocoenose community is applied where the importance is placed on species interdependence. This approach, however, is not applicable to modern marine communities because of the relative in- dependence of marine level-bottom species ( Speden 1966: 411). The habitat community approach of Newell et al. ( 1959: 200) is based on "environmental" species, or the relationship between environmental pa- rameters and certain index taxa. A third kind of definition is based on the classical work of Petersen ( 1911, 1913, .1914, 1915, 1918),in which the most "characteristic" species of animals were used to designate the com- munity. The Petersen communities are synonymous with the organism communities of Newell et al. ( 1959). The Petersen community is based on numerically dominant, recurring species which are designated as the standard-bearers (Thorson 1957: 467). This concept was expanded by Thorson ( 1955), who designated "parallel level-bottom communities." These parallel communities inhabited the same type of substrate in areas scattered geographically (Odum 1971: 340). The Petersen concept for naming communities has been adopted in this investigation. The species compositions of communities are not fixed and bordering communities commonly integrate ( Johnson 1964: 108). This inhomo- geneity of communities is an argument against communities being real units. The disagreement among marine biologists as to the validity of communities being real ecological units is probably the least critical in level-bottom marine communities. Probably the most uniform place in the biosphere is the level-bottom area of the sea ( Thorson 1957: 469). Johnson ( 1964: 108) convincingly sums up the defense: ". . . All species living today do not occur in all possible combinations in natural assem- blages. Rather, we observe around us a finite number of recurring asso- ciations of species. The high recurrence rates of suites of species in modern environments, especially level-bottom marine environments, are probably not chance associations. Thus it seems reasonable to accept level-bottom communities as real or natural entities. 82 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 In a paleoecological study, the recognition of communities is not an end but only the means to an end. It is well documented that the species composition of communities is not constant, and communities also change with time. There is a natural succession during the stages of community development. Species diversity increases during the pioneering stages of succession and will peak and then decline in the climax ( Margalef 1963). The concept of climax is valid; however, documentation of a climax community that would develop in an ecosystem at equilibrium is very tenuous indeed. All systems strive to reach a condition of equi- librium or environmental stability, and the principal limiting factor is time. Dunbar ( 1960), Margalef ( 1968), and Wilson ( 1969) provide excellent accounts of stability evolution in marine environments. The application of the concepts of community succession and sta- bility is difficult in recent benthic marine studies, and these difficulties are compounded when applied to the fossil record. Fortunately, the level-bottom marine communities and their related environments are relatively simple in comparison with complex systems such as rocky in- tertidal, reef, and terrestrial regimes. Upon recognition of the level- bottom benthic communities, it is necessary to interpret their temporal and spatial variations. These variations are reflected by variations in species diversity. Several workers have attempted to explain variations in species diversity between marine communities ( e.g., Buzas and Gib- son 1969, Johnson 1970, Stevens 1971). The most attractive concept at this time is the "stability-time" hypothesis of Sanders ( 1969) . Sanders defines two end members in explaining differences between communities. At one end is the physically controlled, or «physically accommodated," community in which diversity is low, stress is high, and adaptations are controlled by the physical environment. This type of community is not resource-limited ( Valentine 1971: 51). At the opposite end of the spec- trum is the biologically controlled, or "biologically accommodated, com- munity in which diversity is high, stress is low, and adaptations are con- trolled biologically. This type of community is resource-limited. These end-member communities parallel the immature and mature community of Margalef ( 1968), and perhaps the non-interactive, interactive equi- libria model of Wilson ( 1969). Any natural or real community prob- ably falls somewhere between these end points and is an admixture of the two. If local environmental conditions remain nearly constant, the community undergoes succession toward the state of biological accom- modation. If the stresses remain low, a biologically accommodated com- munity evolves; if the stresses remain high, a physically acc:ommodated community develops (Johnson 1970:286). A marine transgressive-stillstand-regressive faunal sequence can be 1975 SHAAK: BRUSH CREEK 83 recognized by the presence or absence of various taxa, taxonomic asso- ciations, inferred feeding types of modes of life, sedimentologic criteria, and geochemical techniques ( Stevens 1971). The transgressive phase, is marked faunally by opportunistic communities composed of relatively few eurytopic species living in rigorous near-shore marginal environ- ments. The stillstand represents the time of maximum marine inunda- tion and supports the most diverse fauna, that of the stable-mature com- munities. Both eurytopic and stenotopic species are found in this higher stability regime. The regressive phase results in a biological winnowing out of the stenotopic species, resulting in relict-mature communities composed of the more eurytopic components of the stable mature phase ( Rollins and Donahue 1971). Certain suites of fossils are recurrent both vertically and laterally, and adequately reflect regional shallow-benthic communities. Detailed paleoecological analyses, including species diversity, equitability, taxo- nomic frequencies, and feedihg and habitat preferences, serve to docu- ment the similarities between the faunal successions at the different localities and warrant the naming of recurrent regional communities. One of the greatest attributes of the "stability-time" hypothesis is that it predicts recognizable changes within marine communities. Spe- cies diversity varies directly with environmental stability ( Donahue and Rollins 1972: 4) and is predictable in a transgressive-stillstand-regressive sequence. The Brush Creek marine transgressive front provided numerous new habitats for colonization by marine organisms ( Rollins and Donahue 1972 : 3). The frontal environments would have been of high stress and fluctuation and would have supported only a few eury'topic species. The communities would have been opportunistic, and in the stability- time framework nearer the physically accommodated end. The species diversity values would be low with the exception of localized "chance" patches of high diversity. As environmental stress decreased or fluctuated less and transgression continued, colonization would increase at a given locality. Stenotopic species would invade the area and physical accommodation would de- crease, whereas biological accommodation would increase. Increased stability would be paralleled by diversity increases. Eventually maxi- mum transgression would be reached and if maintained, stillstand com- munities would develop. The highest stability at a given locality would be reached during the stillstand, and communities would approach bio- logical accommodation, increase in diversity, enjoy low stress environ- ments, and, if time allowed, would continue through succession to ap- proach equilibrium. These stillstand communities would be stable- 84 BULLETIN FLORIDA.STATE MUSEUM Vol. 19, No. 2 mature communities with both stenotopic and eurytopic species, and their successional level or maturity would be a function of time. A high diversity stillstand average for several localities would suggest a rela- tively lengthy period of stability and ecosystem maturity. The regressive phase should show almost the reverse of the trans- gressive phase. Increasing environmental stress and fluctuations would cause a deterioration of the communities. Physical accommodation would become increasingly important as the stenotopic species are biologically winnowed out. The residual communities would be the relict-mature communities composed of the more eurytopic species left over from the degraded stable-mature communities, Theoretically, the relict-mature communities would be numerically represented by diversities intermedi- ate between those of the transgressive and stillstand phases. CHARACTERISTICS OF THE BRUSH CREEK DEPOSITIONAL BASIN The Brush Creek marine depositional basin covered parts of Pennsyl- vania, Maryland, Ohio, Kentucky, and West Virginia. In early Brush Creek time the sea was essentially landlocked ( thus brackish), with only a narrow seaway or strait that was the connection with the major mid- continent Pennsylvanian sea extending through portions of Muskingum, Guernsey, and Noble counties in Ohio. The majority of the terrigenous sediments were being shed from the southeastern bordering Appalachian highlands by a river system that built an extensive deltaic elastic edge, prograding toward the center of the basin ( M6rris 1967: 111). The Brush Creek marine deposition commenced with transgression of the Brush Creek sea from the midcontinent region, through the Ohio straits, and into the Brush Creek depositional basin. Concurrently, along with the transgression, the basin ( originally brackish) took on a marine aspect ( Morris 1967: 2). The study area ( western Pennsylvania ) was that part of the basin bounded by a low delta plain to the east and the pro-delta portion of the basin to the west ( Ferm, pers. comm.). Therefore the Brush Creek basin within the study area was directly influenced by both the prograding deltaic wedge and the transgressing sea. FAUNAL SUCCESSION AND DIVERSITY In order to avoid a cumbersome outcrop-by-outcrop treatment of faunal succession and species diversity, the taxonomic data were grouped and examined to ascertain faunal trends. Total outcrop faunal composi- tion, stillstand composition ( the samples from each locality judged to represent the stillstand), stillstand, sub-stillstand, and supra-stillstand 1975 SHAAK: BRUSH CREEK 85 composition, as well as associated lithologic types are the bases for sub- dividing the total fauna into two gross faunal groups, namely high and low diversity molluscan faunas. Of the nine localities studied ( Fig. 1), the geographically peripheral localities ( 1, 7, 8, and 9) contain high diversity molluscan faunas, where- as the more central localities ( 2,3,4,5, and 6 ) contain low diversity molluscan faunas. Although very general in scale, this grouping is very real and critical in the interpretation of diversity and community struc- tures. Taxonomic frequencies in the samples considered representative of the stillstand average 15 molluscan taxa for the high diversity molluscan faunas, whereas the low diversity molluscan faunas in stillstand samples average 10 molluscan tan ( Table 2). The high diversity faunas average 62% molluscan taxa of the total number of taxa, as opposed to 46% molluscs for the low diversity faunas. Though ubiquitous, Astartella ( a small rhomboidal to ovate eurytopic bivalve of the family Astartidae) is 10 times more abundant in the high density molluscan faunas than in the more central ones. Total outcrop faunal compositions, although less useful than stillstand data, show similar relationships. Molluscs average 15 taxa in the high diversity faunas, whereas the low diversity faunas average 11 molluscan taxa. The molluscan taxa from the high diversity faunas average 61% of the total number of taxa, and the low diversity average is 47% mollus- can taxa. Astartella is four times more abundant in the high diversity molluscan outcrops ( Table 3). These data lead to a suggestion of relative water depth and distance from shore for the depositional environments of these two outcrop groups. Pennsylvanian molluscan faunas in shallow benthic communities in Colorado increase in diversity nearshore according to Stevens ( 1971: 406), whereas other faunal components increase in diversity offshore. Thus the high diversity molluscan faunas of the western Pennsylvania Brush Creek marine unit represent marginal or shoal water localities ( Rollins and Donahue 1972). There is also lithologic support for these conclusions. Locality 7, near Uniontown, Pennsylvania, has its diversity peak at the position of a sideritic ironstone layer containing a molluscan dominant fauna and associated phylloid algae, representing an interdistributary bay facies ( Donahue, Rollins, and Shaak 1972: 8). Similarly, the remaining high diversity molluscan localities ( 1, 8, and 9) have bedded ironstone no- dules below the position of their diversity peaks. In the low diversity molluscan outcrops the first occurrence of ironstone nodules, either bedded or scattered upward from the base of the fossiliferous zone, is TABLE 2.-SUMMARY OF STILLSTAND SAMPLE FREQUENCIES. 86 B U LLE T IN FLO R ID A S TATE M U S E U M Vol. 19, N o. 2 Sample Locality 123456789 Shansiella 13445 18011 3 748 48810 0 0 0 · 0 Plagioglypta 9942 0 106528 0 6894 6190 10398 392 12066 Aclisina 7753 919 3272 444 3705 28378 5218 9100 284 Crinoids 1370 0 81 2271 3148 3228 0 60 981 Ornamented Ostracods 970 0 1879 1174 110 0 298 100 653 Smooth Ostracods 2891 5450 2767 5373 525 3480 5783 1656 6577 Cyclogyra 657745 243071 246744 630958 16652 179599 0 665 967639 Glabrocingulum 180720 2521 275042 19700 1399 221960 108066 7999 92502 Astartella 189335 26173 699 296 7844 312337 312337 10397 93512 Inathanopsis 54936 5304 5602 0 34012 556 8726 5193 13181 Molluscan taxa 19 10 11 10 10 12 13 13 16 Number of taxa 31 23 25 22 21 23 25 18 26 Molluscan % of taxa 61 43 44 45 48 52 52 72 61 TABLE 3.-SUMMARY OF TOTAL OUTCROP TAXONOMIC FREQUENCY AVERAGES. Sample Locality 123456789 Shansiella 2421 23127 3 1732 63242 789 0 29 10 Plagioglypta 985 1 33403 62 2862 4095 12294 87 13604 Adisina 1298 4074 3269 788 2834 36727 870 14485 3618 Crinoids 764 2 361 1107 1560 655 2111 127 341 Ornamented Ostracods 209 159 353 8926 95 199 139 17 92 Smooth Ostracods 3517 3113 1136 8654 917 1523 5161 3296 3106 Cyclogwa 255826 359156 91268 608586 8175 113118 0 612 434907 Glabrocingulum 33373 1452 68000 6812 646 239608 63026 26499 11835 Astartella 36868 46811 984 10692 21971 103480 442384 13845 82243 Ianthanopsis 8623 1246 1152 161 6020 959 11103 46264 33276 Stillstand number of taxa 31 23 25 22 21 23 25 18 26 Molluscan taxa 26 15 11 15 13 17 20 21 20 Number of taxa 46 38 28 31 25 32 39 29 31 Molluscan % of taxa 57 39 39 49 52 53 51 72 65 1975 SHAAK: BRUSH CREEK 87 above the position of the diversity peak; that is, in the regressive sequence ( Appendix A). The significance of the bedded nodules and their rela- tionship to the sideritic ironstone layers at Loc. 6 and the overall rela- tionship to water depth and distance from shore is that many of the aforementioned bedded nodules contain fractures resembling desicca- tion cracks, further suggestive of shoal conditions ( Wilson 1967: 85, Emery 1950: 220). One other taxonomic group provides excellent supportive evidence for a shallow water origin of the high diversity molluscan samples. The ornamented ostracods may require near normal marine salinities for the precipitation of their elaborate shells ( Benson 1961: Q58). Looking at total outcrop taxonomic frequency averages ( Table 3), the ornamented ostracods are 17 times more abundant in the low diversity molluscan outcrops, whereas smooth ostracods are more abundant in the high diversity molluscan localities. Such treatment of data, which was originally intended only for a general grouping of faunas from the nine localities, proves to be a sur- prisingly valuable paleoecological tool in this investigation. It is within the framework of low versus high diversity molluscan faunas that the individual localities are discussed with respect to species diversity and faunal succession. HIGH DIVERSITY MOLLUSCAN FAUNAS LOCALITy 1, KrITANNING, PENNSYLVANIA.-Eleven trench samples, excluding the limestone ( 1.5 ft.), were trenched in an 11 foot interval from 17-28 ft. above the Brush Creek coal. Data derived from the 11 samples show a distinct molluscan dominancy, as evidenced by the number-of-taxa curves ( Fig. 4). The equitability curve shows an in- verse relationship to the molluscan curve. At molluscan lows there are corresponding equitability highs, and molluscan highs are paralleled by equitability lows. This inverse relationship is explained by dominance within the molluscan fauna of the omnipresent bivalve Astartella. The diversity index curves essentially parallel the equitability curve and roughly parallel the species diversity curves. These relationships, as well as the numerical values calculated for these curves, delineate the. transgressive-stillstand-regressive phases of the marine event. The ma- rine faunal stillstand is judged to occur at the position of Sample 8, 1.0- 2.0 ft. above the limestone. This sample contains 31 taxa, of which 19 ( 61%) are molluses. This species diversity high is paralleled by Shan- non-Wiener and Simpson index highs of 2.7 and 4.3 respectively ( Table 4). The equitability values for this sample ( 0.14) is an intermediate feel sample - 28 - - 88 B U LLE T IN FLO R ID A S TATE M U S E U M Vol. 19, N o. 2 11 1 27- - / A10 1 \ 726 - - 9 \ \\ 25- - 00 \. \ 8 5 , 24 - - 7 , r - 6 0/4 - 5 - "b23- - 1,limeslone , \ \ 21- 4 - ijl20- 3 \ \19- - \ 2 T .,18 - - / 1/ 17- - 1,1,11 1 11,1 1 012345 0 3 10 15 20 25 30 0 .1 .2 .3 .4 DIVERSITY INDICES NUMBER OF TAXA EQUITABILITY Shannon - Wiener total fauna Simpson mollus«s only - without molluscs-·- FIGURE 4.-Summary of sample locations, diversity indices, number of taxa, and equitability for Locality 1, Kittaning, Pennsylvania. 1975 SHAAK: BRUSH CREEK 89 value for this locality and again reflects an internal dominance within the molluscs. There are several anomalous peaks on the curves that require an ex- planation. First, diversity index highs ( 2.6 and 4.6) occur within sample 2 ( 2.5-3.5 ft. below the limestone). These highs correspond to an equitability high of 0.29. The explanation is that only 16 taxa are re- corded from the sample and the frequencies are evenly spread among these taxa. Thus the diversity index highs are controlled, in this case by an equitability high. This same factor controls the diversity index peaks in Sample 6. Secondly, at sample 4 there is a marked species diversity high which closely approximates the stillstand values at Sample 8. However the diversity indices are intermediate in value ( 2.3,3.0) at Sample 4, and the corresponding equitability value is low (0.1). This anomalous sample probably reflects patchiness in the areal distribution of the fauna and reflects a local diversity and density high ( garden or oasis) within the transgressive phase that was a "chance" find in the sampling. Such patchiness is evidently more common in both the fossil record and in modern faunas than is generally recognized ( Speden 1966: 413). The assumption that this sample represents a local diversity high is further evidenced by the fact that both species diversity and diversity index values are intermediate in the subjacent and suprajacent samples. This transgressive high represents an opportunistic fauna dominated by eurytopic organisms such as Cyclogyra and Glabrocingulum. The transgressive-regressive phases are distributed asymmetrically above and below the stillstand. The stillstand contains the stable mature fauna. LOCALITY 7, UNIONTOWN, PENNSYLVANIA < PICCOLOMINI LOCALrrY ).- A report on a paleoecological investigation of this locality has been pub- lished ( Donahue, Rollins, and Shaak 1972); only the salient features are mentioned here. Faunal analysis of six trenched samples taken within a three foot continuous interval suggests the following conclusions. First, species diversity, diversity indices, and equitability data ( Fig. 5) suggest that the stillstand occurred within a limestone-ironstone unit and asso- ciated subjacent and suprajacent shales. Secondly, the diversity indices at the stillstand are qualitatively estimated to be 3.0 and 4.0 ( Table 5), with an equitability values of 1.3. The estimates weremade from hand picked specimens of the highly fossiliferous limestone-ironstone that could not be treated with the conventional laboratory techniques ( a discussion of the limestone problem is found at the end of this section). Thirdly, a second diversity peak occurs within Sample 5, which is domi- nated by a productid-chonetid brachiopod fauna with smaller numbers of bivalves. This regressive high represents a somewhat stable relict- leet 14 - 90 B U LLE TIN FLO R ID A S TATE M U SEU M Vol. 19, N o. 2 - sample 6 0/ 1/13 - \ \ - 5 \ \\ 11 12.- 24 \ Tr \ 1\\ - Itmes*one - Ironstone .1 11- -3 • 1,11'11 01 2 3 4 05 10 15 20 25 30 0 .05 .10 .15 10- DIVERSITY INDICES NUMBER OF TAXA EQUITABILITY Shannon -Wiener total fauna - Simpson --- mollus€s only --- without mollus€s - - - FIGURE 5.-Summary of sample locations, diversity indices, number of taxa, and equitability for Locality 7 ( Piccolomini locality), Uniontown, Pennsylvania. 1975 SHAAK: BRUSH CREEK 91 TABLE 4.-SUMMARY OF DIVERsrrY AND EQUrrABILrrY VALUES FOR LOCALGY 1, KrrrANING, PENNSYLVANIA. Sample Number Molluses Without S-Wl SZ Equita- Number Interval of Taxa Only Molluscs Diversity Diversity bility 11 4.0-5.0' above Iimestone 14 8 6 2.04 3.16 .23 10 3.0-4.0' above limestone 15 7 8 2.05 2.93 .20 9 2.0-3.0' above limestone 14 8 6 1.53 2.39 .17 8 1.0-2.0' above limestone 31 19 12 2.74 4.26 . .14 7 8"-1.0' above limestone 15 7 8 1.65 2.27 .15 6 4-8" above limestone 13 6 7 2.48 3.94 .30 5 0-4" above limestone 20 10 10 1.28 1.63 .08 4 0-1.0' below limestone 30 18 12 2.29 3.04 .10 3 1.0-2.5' below limestone 20 13 7 2.06 2.96 .15 2 2.5-3.5' below limestone 16 6 10 2.65 4.57 .29 1 3.5-4.5' below limestone 10 5 5 2.00 3.09 .30 1 S-W-Shannon-Wiener 2 S=Simpson mature fauna dominated by the more eurytopic components left over from the stable-mature ( stillstand) phase. Lastly, the faunal distribu- tion above and below the stillstand is asymmetrical. LocALrry 8, URSINA, PENNSYLVANIA.-This locality is a collector's paradise for an exquisitely rich molluscan fauna. The bellerophontid gastropod Pharkidonatus is very abundant and, because of its large size, is the most conspicuous component of the macrofauna. Seven trench samples were taken within a 6.5 ft. stratigraphic in- terval. The resultant data curves ( Fig. 6) are diagnostic and again show the molluscan dominance characteristic of these shoaler localities. The stillstand phase peaks in Sample 5 at diversity index highs of 2.6 and 5.1 ( Table 6). The equitability value of 0.28 for Sample 5 explains TABLE 5.-SUMMARy OF DIVERSITY AND EQUITABILITY VALUES FOR LOCALrrY 7, UNIONTOWN, PENNSYLVANIA. Sample Number Molluscs Without S-Wl SZ Equita- Number Interval of Taxa Only Molluscs Diversity Diversity bility 6 10.0-18.0" above limestone 15 7 8 1.50 1.68 .11 5 0.5-10.0" above limestone 23 12 11 2.10 2.54 .11 4 0-0.5" above limestone 19 9 10 1.67 2.20 .12 limestone-ironstone 3 0-0.5" below limestone 25 13 12 1.39 1.60 .06 2,10.5-2.0" below limestone 20 10 10 1.05 1.40 .07 1 S-W=Shannon-Wiener 2 S=Simpson the diversity index peaks on the basis that the faunal distribution is evenly spread among the various taxa. This sample contains a stable- fee, 13- Zomple 12- 7 T 11 -\ \ 11- 6 1 14 \\ -- \\\ 1 2 10- - / 11limestone - / 1 B U LLE T IN FLO R ID A S TATE M U S E U M Vol. 19, N o. 2 4 ,9-- 1 // 3 < f f\ \18-- \\ 2 7-- / 1/ il1 6- - 1,1,11 1.,i, 1. 1., 0 1 2 3 4 5 0 5 10 15 20 0 .1 .2 .3 .4 .5 DIVERSITY INDICES NUMBER OF TAXA EQUITABILITY Shannon-Weiner . total fauna - 5- Simpson --- molluscs only --- without molluscs - - FIGURE 6.-Summary of sample locations, diversity indices, number of taxa, and equitability for Locality 8, Ursina, Pennsylvania. 1975 SHAAK: BRUSHCREEK 93 TABLE 6.-SUMMARY OF DIvERSITy AND EQUITABILrrY VALUES FOR LOCALITY 8, URSINA, PENNSYLVANIA. Sample Number Molluscs Without S-Wl S: Equita- Number Interval of Taxa Only Molluses Diversity Diversity bility 7 1.5-2.5' above limestone 4 3 1 0.63 1.26 .32 6 0.5-1.5' above limestone 7 4 3 1.14 1.72 .25 5 0-0.5' above limestone 18 13 5 2.57 5.07 .28 4 0-0.5' below limestone 18 13 5 1.86 2.88 .16 3 0.5-1.5' below limestone 11 8 3 1.63 2.17 .20 2 1.5-2.5' below limestone 12 8 4 2.23 3 39 .28 1 2.5-3.5' below limestone 8 5 3 2.17 3.58 .45 1 S-W=Shannon-Wiener 9 S=Simpson mature fauna. The very high data curve values suggest a relatively high state of biological accommodation. There is perfect agreement be- tween species diversity at the stillstand and Sample 4, and upon initial inspection suggests a relatively lengthy stillstand characterized by two lateral stable-mature faunas separated by a low diversity limestone facies. However the lower species diversity peaks ( Sample 4) have low di- versity index and equitability counterparts. This phenomenon is ex- plained by numerical dominance of several eurytopic taxa within the faunas, and also by the fact that 8 of the 18 taxa have frequencies of 20 or less. Thus, the species diversity high of Sample 4 is interpreted as a relatively stable opportunistic faunal patch that developed in the late stages of the transgressions. This fauna perhaps gave rise directly to the stable-mature fauna of the stillstand by a decrease in the fre- quencies of the eurytopic components of the opportunistic fauna. The limestone unit subjacent to the stillstand position is a trans- gressive limestone. A qualitative analysis indicates that its fauna is quite low in diversity and probably represents a restricted environment in which a limy mud was deposited under brackish conditions. The faunal succession is markedly asymmetrical above and below the stillstand. The high diversity index values at the stillstand suggest a relatively high level of biological accommodation. Locurry 9, GLADE CITY, PENNSYLVANIA.-This locality is unique in two aspects. First, two limestones are exposed in the sequence. The upper limestone is clearly of channel origin, and it occures as a lens- shaped body that grades laterally in two directions into bedded ironstone nodules. The channel is 5 feet wide and 0.5 feet in maximum thickness. Lithologically, the limestone is very dense, very black, and nearly devoid of fossils. This limestone perhaps was deposited as a black mud in a restricted abandoned channel on the lower part of a delta plain. Second, 94 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 the number of taxa curves graphically indicate the faunal succession ( Fig. 7). Eight trench samples were taken within a 7.5 foot stratigraphic interval. The resultant species diversity curves unquestionably pinpoint the marine stillstand. Molluscs again control the faunal succession, as shown by the parallelism of the total fauna and mollusc curves and the opposite relationship between the mollusc and non-mollusc curves. The diversity index curves are rather nondescript and, used alone, are mis- leading. The stillstand Occurs within Sample 5. The diversity index curves show a slight infiection at the stillstand, which is in agreement with a low equitability value. These low values at the stillstand are explained by an uneven taxonomic frequency spread caused by the dominance of Cyclogyra, a planispiral foraminifer of the family Fischer- inidae, and Tol!/pammina, a tubular encrusting foraminifer of the family Ammodiscidae. There is a diversity index high and corresponding equitability high at the Sample 2 position, which is readily explained by the presence of only seven taxa that are evenly distributed. This equitable frequency distribution among the seven taxa pumps up the diversity indices regard- less of the number of taxa present ( Table 7). TABLE 7.-SUMMARY OF DIVERSITY AND EQUHABILKY VALUES FOR LocALrrY 9, CLADE CITY, PENNSYLVANIA. Sample NumberMolluscs Without S-Wl Equita- Number Interval of Taxa Only Molluses Diversity Diversity biaty 8 2.5-3.5' above limestone 6 4 2 1.42 221 .37 7 1.5-2.5' above limestone 16 7 9 2.05 2.81 .18 6 10.0-14.0" above limestone 24 13 11 2.05 2.63 .11 5 5.0-10.0" above limestone 26 16 10 2.31 2.84 .11 4 0-5.0" above limestone 23 13 10 2.06 2.43 .11 3 0-1.0' below limestone 9 4 5 1.67 2.38 .27 2 1.0-2.0' below limestone 7 4 3 1.99 3.30 .47 1 2.0-3.0' below limestone 5 3 2 1.04 1.96 .39 1 S-W=Shannon-Wiener B S=Simpson The faunal succession is markedly asymmetrical with a gradual transgressive phase composed of a few eurytopic taxa, peaking at the stillstand, and degrading relatively rapidly in the regressive phase. Qualitatively, the limestone is of low diversity and probably was de- posited in a brackish restricted environment. The limestone was formed during transgression. feet 12 - 1975 S H A A K : BR U SH C R E E K sample 11 - 8 \\\ 10 - \ \\ 7 i <\~\ 1 \ 1 4 9- - 6 4 \ 5 ) 1 N/8- limesione 1,- 1 7- \ 6 - \ .12 l /1/ li / 11 5- 1 4- , , 1,,1 1 1 01234051015202530 1 .2 .3 .4 .5 .6 DIVERSITY INDICES NUMBER OF TAXA EQUITABILITY Shannon -Weiner - total fauna - Simpson -- - mollusci only --- without molluscs - - a FIGURE 7,-Summary of sample locations, diversity indices, number of taxa, and equitability for Locality 9, Glade City, Pennsylvania. seel sample 14 - - 96 B U LLE TIN FLO R ID A STATE M U SEU M Vol. 19, N o. 2 5 13- - 1 il 12- 4 \ - \ \1 11- 3 \; It- / 1\ 10- limestone / 1- // i \1/ \9- 2 . \\ i /\\ i /8- \\1 / il 7- 6i23465 10 1520256 .65 .10 .15 .20 6- DIVERSITY INDICES NUMBER OF TAXA EQUITABILITY Shannon-Weiner - total fauna - Simpson --- mollus€s only --- withoul mollus€s - - - FIGURE 8.-Summary of sample locations, diversity indices, number of taxa, and equitability for Locality 2, Shelocta, Pennsylvania. 1975 SHAAK: BRUSH CREEK 97 Low DIVERSITY MOLLUSCAN FAUNAS LOCALITY 2, SHELOCrA, PENNSYLVANIA.-Brant ( 1971) presented a very detailed interpretation of the faunal succession of this locality. He used species diversity, equitability, and geochemical data for the pur- pose of zoning the section into communities. Brant defined 12 zones, of which zones 1 through 6 were classified as transgressive and zones 11 and 12 regressive. He did not indicate the position of the stillstand, but stated: "the position of the stillstand in this section does not appear meaningful" ( Brant 1971:72). Although the data curves for this locality ( Fig. 8) are relatively flat, two diversity index highs are shown. These curves have a near- perfect fit with the equitability curve which, is explained by an "even- ness" of spread of the frequencies within the various taxa for the highs and a d6minance factor for the lows. The stillstand is judged to occur within Sample 3. Molluses com- prise 43 percent of the total number of taxa. Absolute high diversity index values of 2.5 and 3.1 also occur within this sample ( Table 8). TABLE 8.-SUMMARY OF DIVERSrrY AND EQUrrABILMY VALUES FOR LOCALITY 2, SHELOCTA, PENNSYLVANIA. Sample Number Molluscs Without S-Wl S2 Equita- Numb3r Interval of Taxa Only Molluscs Diversity Diversity bility 5 2.0-3.5' above limestone 22 10 12. 1.78 2.09 .095 4 1.0-2.0' above limestone 21 8 13 1.83 2.29 .11 3 0-1.0' above limestone 23 10 13 2.46 3.13 .14 2 0-1.0' below limestone 21 9 15 1.39 1.70 .07 1. 1.0-3.0' below limestone 20 8 12 2.16 2.76 .14 1 S.W=Shannon-Wiener 2 S=Simpson, The diversity high at Sample 1 may represent a transgressive or oppor- tunistic faunal patch that is controlled by eurytopic non-molluses, in- cluding 10 taxa ( 50 percent) of foraminifers. Although surprisingly mature for an opportunistic fauna, Sample 1 does not reach the succes- sional maturity of Sample 3 ( stillstand). These observations indicate that diversity index data alone can be misleading. The reBnement of the raw faunal assemblage data of the "reflection" of standing crop provides the basis for the interpretation of this outcrop. The limestone is qualitatively judged to be of low diversity, con- taining Kegelites ( an amphissitid ostracod with a pitted shell) and chonetid brachiopods. Brant ( 1971: 52) gave a Shannon-Wiener aver- age index for the limestone of 1.25, which is in agreement with a quali- -eet 11 - 98 B U LLE TIN FLO R ID A S TATE M U S E U M Vol. 19, N o. 2 sEmple 10 - 5 \ \ 9- 4 1 3 ( limestone) 0 \ \ 2 .jl /1 7- - 1 • lili 1 1 1 1 1 1 1 1 1 1 ! 1 6- 0 1 2 3 4 0 5 10 15 20 25 .1 .2 .3 .4 .5 .6 DIVERSITY INDICES NUMBER OF TAXA EQUITABILITY Shannon -Wiener - total fauna - Simpson --- molluscs only --- without mollusts - - - FIGURE 9.-Summary of sample locations, diversity indices, number of ttaxa, and equitability for Locality 3, Barton, Pennsylvania. 1975 SHAAK: BRUSH CREEK 99 tative guess of low diversity in this study. However, he claimed that the limestone is diagenetic ( secondary ) in origin ( Brant 1971: 9), based on its variable thickness and the localized gradational contact with the subjacent and suprajacent shales. This interpretation is in disagreement with that of the present study. The Brush Creek limestone is laterally persistent and in several localities ( 3 and 7) contains a stable mature fauna. Local lithologic variations, such as those at this locality, refect different limestone lithotopes within the various limestone facies. The limestone at this locality is interpreted as a transgressive limestone de- posited in a marginal but restricted environment. The restricted nature is evidenced by the absence of stenotopic organisms. The transgressive-regressive phases are asymmetrically distributed above and below the stillstand. LocALrrY 3, BARTON, PENNSYLVANIA.-Five trench samples were taken within a 7.5 foot stratigraphic interval. The diversity index curves ( Fig. 9) for this locality are controlled by extreme fluctuations in equi- tability. The stillstand is judged to occur within Sample 3, the limestone unit, as evidenced by the species diversity curves. The lack of agree- ment between the diversity index curves and species diversity curves at the level of the stillstand is explained by the corresponding equitability low ( 0.15). This equitability low is a function of dominance by two taxa, Cyclogyra and Glabrocingulum ( the latter a small pleurotomari- acean gastropod). The stillstand, or stable-mature, fauna represents a relatively short stillstand duration, as evidenced by the lack of an inflec- tion in the diversity index curves. The stillstand fauna failed to reach a high level of ecosystem maturity or biological accommodation even though the number of taxa increased to an absolute high of 25 ( Table 9). TABLE 9.-SUMMARY OF DIVERSITY AND EQUITABILITY VALUES FOR LOCALITY 3, BARTON, PENNSYLVANIA. Sample Number Molluscs Without S-Wl Se Equita- Number Interval of Taxa Only Molluscs Diversity Diversity bility 5 0.5-1.5' above limestone 8 3 5 0.86 1.43 .18 4 0-0.5' above limestone 18 6 12 2.37 3.73 .21 3 limestone 25 11 14 2.18 3.65 .15 2 0-0.5' below limestone 8 2 6 2.09 3.50 .44 1 0.5-1.5' below limestone 3 1 2 1.00 1.63 .54 1 S-W=Shannon-Wiener 2 S=Simpson Because it contains a stable-mature fauna, the limestone at this 10- cality is unique. Lithologically the limestone is very argillaceous and poorly cemented, thus very friable. The fossils are exquisitely preserved 100 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 and readily removed from the matrix by the Amine 220 technique. This particular limestone perhaps represents a true pro-delta basinal organic limestone with minor but continuous terrigenous influx. The terrigenous fraction is probably the cause of the friable nature of the limestone and perhaps was the limiting factor in the low level of biological accommoda- tion. The transgressive-regressive phases are distributed asymmetrically around the stillstand, with a relatively rapid transgressive faunal bloom peaking at the stillstand. The regressive phase is less pronounced, with a gradual biological winnowing of the stenotopic components of the weakly developed stable-mature fauna. LOCALrry 4, SEWICKLEY, PENNSYLVANIA.-Six trench samples were taken within a 5.5 ft. interval, including the lower 0.5 ft. of the 1.5 ft. thick limestone unit ( Table 10). The resultant data and associated TABLE 10.-SUMMARY OF DIVERSITY AND EQUrrABILITY VALUES FOR LOCALITY 4, SEWICKLEY, PENNSYLVANIA. Sample Number Molluscs Without S-Wl 52 Equita- Number Interval of Taxa Only Molluscs Diversity Diversity bility 6 1.0-2.0' above limestone 14 7 7 1.12 1.68 .12 5 0.5-1.0' above limestone 22 10 12 1.25 1.64 .07 4 0-0.5' above limestone 17 5 12 1.20 1.58 .09 3 lower 6.5' of limestone 19 10 9 1.10 1.59 .08 2 0-1.0' below Iimestone 19 11 8 0.86 1.33 .07 1 1.0-2.0' below limestone 19 10 9 1.45 1.79 .09 1 S.W-Shannon-Wiener 9 S=Simpson diversity curves ( Fig. 10) at Brst appearance were rather non-descript and difficult to interpret. However several subtle faunal trends pro- vided an insight into the faunal succession. The values along all of the data curves are quite low compared with the other localities. The equitability scale has been exaggerated to show inflections that normally would be masked at a more reduced scale. The diversity index curves deviate only slightly from a straight line owing to a marked dominance in all samples by two foraminifer genera, Cyclogga and Tolypammina. The successional interpretation is derived for this locality from data other than the diversity index curves. The key to the interpretation is in the inverse relationship between the molluscs and non-molluscs in the number of tan or species diversity curves ( Fig. 10). Going up- section, the number of molluscan taxa is greater than the number of non- molluscan taxa, Within the limestone the two curves intersect and the feet 7- 1975 SH AAK: B R U S H C R E E K 101 Zomple 6-6 f T 1. - 1 \\1 >, 5.I - 1 // i 4 \. ! 4 - linestone \/ - 3 1 1\ 3- - V / i 1 2 \ \2- - \ 1 \ .1\ 11 1 1- - ,6 .5 £0 1.5 2.0 6 5 10 15 20 25 0 .05 .06 .07 .08 .09 .10 .11 .12 DIVERSITY INDICES NUMBER OF TAXA EQUITABILITY Shannon -Wiener - total fauna Simpson --- mollus€s only --- without molluses -- FIGURE 10.-Summary of sample locations, diversity indices, number of taxa, and equitability for Locality 4, Sdwickley, Pennsylvania. 102 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 non-molluscan taxa become dominant, suggestive of an increase in water depth ( Stevens 1971 ). A molluscan low and a non-molluscan high are reached within Sample 4. This probably represents the maximum trans- gression ( thus deepest water) at this locality. These data correspond to an equitability high of .09. The foraminiferal dominance is still para- mount; however Sample 4 is more equitable than the subjacent and suprajacent samples. The ornamented ostracods and the crinoids reach an absolute high for the outcrop within Sample 4. Astartella reaches an absolute low in this sample. This is all suggestive of deeper water. Thus, the stillstand is judged to have occurred within Sample 4, based on relative water depths. On visual inspection only, the species diversity curves would indicate that the stillstand occurred within Sample 5. The key to this interpreta- tion is again the molluse and non-mollusc relationship. The molluscs reach a taxonomic frequency low within Sample 4 and increase from 5 to 10 taxa ( Table 10) at Sample 5. The non-molluscan number of taxa remains unchanged from Sample 4 to 5 ( 12 taxa). This phenomenon re- Rects a molluscan diversity pumping resulting from shoaling effects. Even though five molluscan taxa are added to the fauna, their frequen- cies are so low that with the increase of the dominant foraminifers ( be- cause of shoaling) the equitability value decreases drastically. The reduction of ornamented ostracods by a factor of five and the threefold increase of Astartella are further suggestive of shoaling conditions ( Fig. 10). Thus Sample 5 reflects a very diverse relict-mature fauna. There is an alternative explanation for the increase in the number of taxa within Sample 5. The fauna of Sample 5 may reflect an ecotone or a transition between two communities. The increase in the number of taxa from Sample 4 to 5 is entirely molluscan. This increase might be explained by the "edge effect" ( Odum 1971: 1957) in which there is often a greater faunal diversity and density in the ecotone than in the adjoining communities. This edge effect may represent the junction zone between the early relict-mature fauna and an adjacent shoal water mol- luscan fauna. Unfortunately, this concept cannot be documented with the data available from this investigation. There is another peculiar attribute of this locality that warrants an explanation. The lower .75 ft, of limestone- is so poorly consolidated and friable that the fossils were removed from the matrix with ease and treated with the standard data recovery techniques. The upper .75 ft. of limestone is argillaceous, but less Mable and more highly consolidated than the lower .75 ft. This suggests increased terrigenous influx during the deposition of the upper segment of the lime, and might explain the decrease in the number of molluses through the limestone, peaking at a 1975 SHAAK: BRUSH CREEK 103 low in Sample 4. The increased terrigenous influx would have affected the various feeding types ( esp. filter feeders) and could have caused a decline in the number of taxa ( Purdy 1964: 243). However the maxi- mum density of the crinoids and fenestrate ectoprocts within this sample negates any strong sedimentologic inf[uence. The distribution of the transgressive-regressive phases around the stillstand is asymmetrical within the confines of the sampling design. Unfortunately the initial taxonomic buildup of the transgressive phase is lower in the section than the sampled interval. The highest and low- est samples have relatively high equitability values, as is common in early transgressive and late regressive phases. LOCALITY 5, GLENSHAW, PENNSYLVANIA.-This locality is unique be- cause it shows an extremely rapid transgression to the stillstand from non-marine, or near non-marine conditions within a stratigraphic interval of only 13 inches (Fig. 11). Six trench samples were taken within a 6.0 ft. stratigraphic interval. The basal unit of the sequence is a massive black mudstone at least 10 ft. thick ( the basal coal is not exposed), which is capped by a six-inch bed of limestone. The sample interval extended two feet into the mudstone. Only two taxa of ostracods ( one smodth and one ornamented) were recovered from the matrix. A second outcrop sevetal hundred yards to the southwest also is barren of fossils below the limestone and lithologically is composed of interbedded shale and ironstone nodules. The black mudstone represents a restricted en- vironment at the extreme end of the physical accommodation and mem- ber of Sanders' stability-time hypothesis. The physiological stress condi- ti6ns were so great that the resultant environments were beyond the adaptive means of most marine animal taxa. The marine transgression was initiated with the deposition of the limestone which, on a qualitative inspection, is very low in diversity. The diversity index and species diversity curves peak within Sample 3 ( Fig. 11), indicating the stillstand. The stillstand peak occurs within 13 inches of stratigraphic section, thus the transgression was nearly in- stantaneous geologically. The high diversity index value of 6.0 for the Simpson equation is a result of the rapid increase in the number of taxa from 1 in Sample 2 to 21 in Sample 3 ( Table 11). These peaks corre- spond to an intermediate equitability Value of 0.28, which is intermediate because of dominance by Shansiella, a large turbiniform pleurotomaria- cean gastropod. The species diversity curves show a second inflection ( Sample 5) in the regressive or relict-mature fauna. The mollusc and non-mollusc curves are nearly parallel through Sample 4, but in Sample 5 the curves feet 14 - sample 104 B U LLE TIN FLO R ID A S TATE M U S E U M Vol. 19, N o. 2 6 , 13- / \1 \ / 1\ 5 / 1\ 1 \112 - 1 1 4 \\ - 11- 3 7 / limeslone / 11 / 10- / 4/ /2 \ 9- \ \ 1 \' '1'',1,1 8-012 3 4 5 6 0 5 10 15 20 25 0 .25 .50 .75 1.0 DIVERSITY INDICES NUMBER OF TAXA EQUITABILITY Shannon-Wiener- Iotal fauna - Simpson --- molluscs only --- without mollus€s -·- FIGURE 11.-Summary of sample locations, diversity indices, number of taxa, and equitability for Locality 5, Glenshaw, Pennsylvania. 1975 SHAAK: BRUSH CREEK 105 TABLE 11.-SUMMARY OF DIVERSMY AND EQUITABILrrY VALUES FOR LOCALrry 5, GLENSHAW, PENNSYLVANIA. Sample Number Molluscs Without S-Wl SZ Equita- Number Interval of Taxa Only Molluses Diversity Diversity bility 6 2.0-3.0' above limestone 8 3 5 2.39 4.91 .61 5 1.0-2.0' above limestone 15 9 6 2.26 3.86 .26 4 0.5-1.0' above limestone 16 7 9 2.29 2.80 .18 3 0-0.5' above limestone 21 10 11 3.03 5.96 .28 2 0-1.0' below limestone 1 0 1 0.00 1.00 1.00 1 1.0-2.0' below limestone 2 0 2 1.00 2.00 1.00 1 S-W-Shannon-Wiener 9 S=Simpson have reversed position, with the molluscs becoming dominant with re- spect to the number of taxa for the first time in the sequence. This molluscan increase corresponds to an increase in equitability and di- versity index values, especially the Simpson function. An abnormally high Simpson index of 4.9 occurs in Sample 6, late in the regressive phase or relict mature fauna. This peak is not controlled by the number of taxa, but by an equitability high (.51) that reflects an even distribution of the frequencies within the eight recorded taxa. The distribution of the transgressive-regressive sequence around the stillstand is markedly asymmetrical, with a rapid transgressive phase and an accompanying rapid buildup of the opportunistic fauna. The regres- sive phase was more irregular than the transgressive phase. The still- stand and accompanying stable-mature fauna was relatively shortlived, as the number of taxa increased, peaked, and decreased within a seven- inch stratigraphic interval. The lack of agreement between the lower portion of the diversity index curves and the equitability curve warrants an explanation. Samples 1 and 2 have a value of 1.0 perfect equitability. However, the corre- sponding diversity index values of the Shannon-Wiener value is 0.0, compared with the perfect equitability of 1.0. The Shannon-Wiener function is strongly influenced by equitability, and in this case the di- versity index should be high. However a perfectly equitable distribu- tion ( one taxon) represents a totally inequitable distribution in the Shannon-Wiener function because of total ( 100 percent) dominance. Thus, because of total dominance, the Shannon-Wiener index is zero. The corresponding Simpson value is 1.0 which, except for infinity, is the lowest calculable value for the equation, again because of total domi- nance. LOCALITY 6, MURRYSVILLE, PENNSYLVANIA.-Six trench samples were taken within a 5.5 ft. stratigraphic interval. The diversity index curves 106 B U LLE TIN FLO R ID A S TATE M U SEU M Vol. 19, N o. 2 feei 7- fmple 6- 6 - . 5- 5 46/\ \\4- / 4 f i -\limesione3- 3€hannel) 1 1., 3 \~X -/ i'2- -2 \ tmetene 1 - - 1 - 1 il 0123456~6s 101520250 .~ .2 .3 .4 .5 .6 .7 DIVERSITY INDICES NUMBER OF TAXA EQUITABILITY O- Shannon-Weiner - total fauna - Simpson --- molluscs only --- without molluscs - - FIGURE 12.-Summary of sample locations, diversity indices, number of taxa, and equitability for Locality 6, Murrysville, Pennsylvania. 1975 SHAAK: BRUSH CREEK 107 TABLE 12,-SUMMARY OF DIVERSrrY AND EQUITABILrrY VALUES FOR LOCA~rry 6, MURRYSVILLE, PENNSYLVANIA. Sample Number Molluscs Without S-Wi S2 Equita- Number Interval of Taxa Only Molluscs Diversity Diversity bility 6 2.0-3.0' above limestonea 5 4 1 1.55 2.77 .55 5 1.0-2.0' above limestone 9 4 5 2.35 4.36 .49 4 0-1.0' above limestone 15 7 8 1.93 2.84 .19 3 0-7.5" below limestone 23 12 11 2.16 3.21 .14 2 7.5-15" below limestone 23 12 11 2.62 4.62 .19 1 21.0-27.0" below limestone 10 6 4 2.91 6.44 .64 1 S-W=Shannon-Wiener 2 S-Simbson 3 =datum-channel limestone ( Fig. 12) alone are meaningless in the interpretation of species diversity and faunal succession. The diversity index curves peak in two samples (1 and 5), both of which are relatively equitable samples representing no marked dominance by any one taxon. Sample 1 is represented by 10 taxa, with a very high Simpson index of 6.4 ( Table 12). As previously explained, the equitability factor is derived by diving the Simpson index ( 6.4) by the number of taxa ( 10), resulting in an equitability index of .64. The equitability value for this sample is an excellent guide in the understanding and interpretation of diversity-equitability data in general. Knowing what the equitability value represents allows one to look ob- jectively at the diversity values. The Simpson index of 6.4 is controlled either by number of taxa ( species diversity) or equitability. Also, the question of adequacy of sample size must be taken into consideration. The data from Sample 1 clearly define this relationship. If the Simpson index is compared with the equitability value ( 6.4 versus 0.64) within the framework of sample size adequacy test as described earlier, the values minus decimal are equal ( in this case to five digits). If the sample-size test is valid, this sample is adequate, removing the question of incomplete data retrieval. Knowing that the sample size is adequate, the diversity index curves can be interpreted. Samp16 1 contains only 10 taxa, thus the diversity index value is not controlled by the number of taxa but by the evenness of spread of the frequencies within the 10 taxa. Again, the high equitability value of Sample 1 controls the high diversity index value. An almost identical condition occurs within Sample 5, where the diversity index peak is a function of high equitability rather than the number of taxa. Sample 1 was taken through a six-inch carbonaceous shale interval between the basal coal and the overlaying limestone ( Appendix A). There is no non-marine sequence on top of the coal as 10 taxa were ~ accounted for within this basal sample. The overlying limestone is quantitatively low in diversity and represents a transgressive limestone. 108 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 The stillstand occurs within Sample 2, a 7.5 inch interval immediately suprajacent to the limestone. Samples 2 and 3 have identical number of taxa values, and trends in the faunal succession. Sample 2 has a higher equitability value and correspondingly higher diversity index values than does Sample 3. Sample 2 has a more equitable frequency distribution among its taxa than Sample 3, which fits better within the definition of a stable-mature fauna. If Sample 2 represents a stillstand and Sample 3 an early regressive phase, Sample 3 should reflect shoaling trends. This is the case, especially in faunal composition. One of the more reliable shoal-water indicators is Astartella, which increases by a factor of 14 ( 36,115 to 489,365) from Sample 2 to Sample 3. Of 19 molluscan taxa recognized from Samples 2 and 3, eight taxa are common to the two samples. Six of these taxa increase in frequency in Sample 3; a further suggestion of shoaling conditions, or initiation of tbe regressive phase. The number-of-taxa curves graphically depict the asymmetrical dis- tribution of the transgressive-regressive phases around the stillstand. The transgressive phase was of relatively short duration, with a rapid buildup of the opportunistic fauna. The stillstand was achieved in Sample 2, with the accompanying degradation of the stable-mature fauna quite regular with respect to the transgressive phase. A channel limestone is exposed at this locality ( Appendix A), which on general inspection is devoid of fossils. Although very interesting with respect to sedimentation and depositional environments, the channeling removed six inches of the regressive phase deposits and associated fossils from the section. The channel has masked another important relation- ship in which there is a reversal in the number-of-taxa-curves from a molluscan to a non-molluscan dominance within Samples 4 and 5. With shoaling conditions resulting from regression of the sea, the molluscs should remain dominant with respect to the number of taxa. This non- molluscan dominance is only of the order of one taxon in each of the two samples and is judged to. be relatively insignificant as the trend re- verts to a molluscan dominance in the highest sample ( Sample 6). THE LIMESTONE PROBLEM All nine outcrops were trench sampled, and the total samples were broken down by the Amine 220 technique. All of the washed residues were split with a standard sample splitter. Thus, from outcrop to micro- scope, these data are sound with respect to randomness. Every fossil within the samples had a chance of being identified and included in the frequency counts. The limestone members of the Brush Creek interval present a worri- some drawback to an otherwise credible data recovery system. The 1975 SHAAK: BRUSH CREEK 109 limestones are not amenable to the Amine 220 technique. Several lime- stone samples were reduced in a jaw crusher and subsequently treated with the Amine 220 technique, but the sample pieces fractured across fossils and only fossil fragments were recovered in the washed residues. Possible alternatives to the system are qualitative analysis of either whole-hand samples or polished slabs. Some hand samples have whole fossils standing out in positive relief, which allows taxonomic identifica- tion but cannot provide adequate frequency data. Polished sections record all taxa in contact with the plane of the cut. Unfortunately the fossils may have any orientation with respect to the plane, thus taxo- nomic identiftcations are most difilcult and very time consuming. The limestone problem was alleviated somewhat by comparison of the various limestone lithologies. Seven of the nine localities have well consolidated limestones that are unbreakable by the Amine technique. The remaining two localities ( Loc. 3, Barton, Pennsylvania; and Loc. 4, Sewickley, Pennsylvania) have poorly consolidated, Mable limestones from which the fossils were readily retrieved. The entire limestone unit from the Barton locality was broken down and quantitatively is judged to represent the stillstand. The lower 0.5 ft. interval of the 1.5 ft. lime- stone from the Sewickley locality broke down and quantitatively is in- termediate in values between those of the subjacent and suprajacent samples within the transgressive phase. The limestones at the remaining localities, with the exception of the ironstone-limestone horizon at the Uniontown locality, are well consolidated and judged qualitatively to be of low diversity. The ironstone-limestone interval at the Uniontown locality is inferred to represent the standstill on the basis of a subjective quantitative analysis of the well preserved and very rich molluscan fauna that was retrieved from well-weathered float samples. Also, the esti- mated diversity index and number of taxa values are substantially greater than in the sub- and suprajacent samples. Thus there appears to be a trend within the limestones, whether real or diagenetic, of an increase in diversity with a decrease in the level of consolidation. A second trend is perhaps more obvious. The limestone unit at two of the nine localities is judged to contain the stable-mature fauna, repre- sentative of the marine stillstand. The limestone unit at the remaining seven localities is judged to contain a segment of the opportunistic fauna, representative of the transgressive phase. At no locality studied in this investigation does the limestone occur above the stillstand position. Two of the localities have the limestone unit between two quantitatively analyzed samples within the transgressive phase and qualitatively sbow corresponding intermediate number of taxa and diversity index values. The remaining seven localities have the limestone subjacent to the still- 110 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 stand samples and qualitatively appear to be in accord with the pre- dicted intermediate values. These two trendi, though subjective, are considered clear enough as to not invalidate the placement of the stillstand at the seven localities where well-consolidated limestones occur. COMMUNHY STRUCTURE GENERAL STATEMENT One of the most consistent features of shallow benthic marine envi- ronments is the recurrence of a particular taxon or taxonomic assemblage. These ubiquitous taxa may be numerically dominant and may reflect some observable physical environmental parameter. If these ubiquitous, numerically dominant, and environmentally sensitive taxa are found in association with other commonly recurrent species or taxonomic groups, the. definition of a fossil community is met. This definition is in accord with the organism and habitat communities of Newell et al. ( 1959) and the marine level-bottom communities within the framework of marine biology ( Bretsky 1969: 46). The recurrence of species or taxonomic groups is explained by inter- specific associations, similar responses to the physical environment, or chance (Johnson 1964: 128). The chance associations are probably minimal in level-bottom communities and most commonly are found among transient species ( planktonic or nektonic) that are relatively un- important at the organizational level of any particular benthic com- munity. It is commonly accepted that benthic species are quite inde- pendent, and the resultant communities are associations of species living together in harmony with the physical environment. The simple organ- izational level of shallow benthic communities is more amenable to a community structure analysis than the more diverse communities, such as coral reef and rocky intertidal regimes. The . most common community analysis tools have been correlation coefficients and factor anaiysis of a data matrix. A data matrix ( Ap- pendix B) was prepared for this study to show either presence or ab- sence of each recognized taxon at all nine localities. Surprisingly, only eight taxa ( Table 13) were found to be ubiquitous within the study area. The faunal succession was traced through the transgressive-still- stand-regressive phases, and a relationship between these phases and the community structure seemed more logical.than taxonomic groupings based on numerical dominance. Thus the transgressive, stillstand, and regressive sample-adjusted frequencies were averaged for the eight ubi- 1975 SHAAK: BRUSH CREEK 111 TABLE 13.-SUMMARY OF TRANGRESSIVE-STILLSTAND-REGRESSIVE SAMPLE FREQUENCY AVERAGES FOR TAXA COSMOPOLITAN TO THE STUDY AREA. Taxa Transgressivel Stillstandz Regressive3 Ostracods ( smooth) 3954 3834 2882 Ostracods ( ornamented) 98 600 389 Astartelta 90507 75190 85065 Clabrocingulum 21904 92502 63829 Ianthanopsis 17040 13181 2189 Aclisina 5714 6564 8968 Plagiogit/pta 1679 16934 8673 Crihoids 847 1238 423 1 Transgressive values are based on an average of 28 transgressive samples. 2 Stillstand values are based on an average of 9 stillstand samples. 3 Regressive values are based on an average of 24 regressive samples. quitous taxa to ascertain community trends in parallel with the faunal succession. If tbe marine interval can be divided into transgressive, stillstand, and regressive phases, the corresponding taxa should be sub- divisible into transgressive, stillstand, and regressive communities. The frequency averages for the eight ubiquitous taxa adequately support this premise. Astartella-Ianthanopsis CoMMuNrry Astartella and Ianthanopsis are selected to serve as name bearers for the transgressive or opportunistic community. Astartella is a small eury- topic crassatellacean bivalve that was an infaunal non-siphonate suspen- sion feeder. It is classified as a "filterer-A" within the framework of the Russian classificatory scheme for marine benthic invertebrates ( Turpaeva 1957: 137). The successional trend of Astartella ( Table 13) is quite unique because it peaks numerically within the transgressive phase, decreases in the stillstand, and increases within the regressive phase. The frequency low within the stillstand phase is predictable because of the shallow water preference of this species. Its regressive bloom ap- proaches, but does not reach, the numerical dominance developed within the transgressive phase. These marine stage frequencies of Astartella lead one to speculate on the successional maturity of the stillstand or stable-mature fauna for the study area as a whole. Although the still- stand frequencies are reduced, they are still relatively high. The still- Stand values for Astartella, as well as any eurytopic organism, would continue to be reduced with more pronounced biological accommodation ( Levinton 1970). However the modest frequency deduction in the still- stand phase suggests a relatively low level of community maturity and an accompanying low level of biological accommodation. IIi terms of the stability-time hypothesis, this trend is also suggestive of an average short stillstand duration. 112 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 Ianthanopsis, the second name bearer of the opportunistic com- munity, was a globular to high-spired fusiform gastropod of the family Subulitidae. It was an epifaunal vagrant browser and deposit feeder of the "collector" group of marine benthic invertebrates ( Turpaeva 1957). This taxon peaks numerically within the transgressive or oppor- tunistic phase and reaches a low within the regressive or relict-mature phase ( Table 13). Thus, Ianthanopsis is a eurytopic component of the community, and after transgressive bloom never again regained its nu- merical dominance. The trend of Ianthanopsis clearly delimits this species as opportunistic, even more so than Astartella. Further suggestive of a natural transgressive community among the ubiquitous taxa are the unornamented ostracods, which also peak numerically within the trans- gressive phase and decline through the stillstand and regressive phases. The Astartella-Ianthanopsis community is dominated by epifaunal and infaunal vagrant deposit feeding organisms ("collectors" of the Tur- paeva classiBcation). This mode of life dominance in conjunction with the paucity of sessile benthonic filterers ("filterers A and B" of Tur- paeva), such as crinoids and bryozoans, is suggestive of an unstable substrate in a high energy regime in shoal water environments. These conditions are predictable for the opportunistic community as the fauna is the product of the extremes of physical accommodation, Glabrocingulum-Plagioglypta COMMUNITY Glabrocingulum is a turbiniform, gradate to conically-spired gas- tropod of the family Eotomariidae. Plagioglypta is a scaphopod of the family Dentaliidae marked by distinct circular wrinkles generally throughout the shell. These two genera are selected as standard bearers for the stable-mature, or stillstand community because of their cosmo- politan nature and their frequency trends through the faunal succession. Glabrocingulum was a eurytopic vagrant epifaunal browser and de- posit feeder of the "collector" benthic group of Turpaeva. The eurytopic nature of this taxon is out of phase with the stable mature concept of community succession. The eurytopic components theoretically are re- duced in frequency, if not entirely removed, as the fauna trends toward biological accommodation. With increasing biological accommodation, the stenotopic components trend toward a state of equilibrium. If the succession goes unchecked as a result of environmental stability over a lengthy stillstand, the opportunistic species would be relatively rare ( Levinton 1970: 69). However, because of the relatively low level of successional maturity and relatively short duration 6f the stillstand, the stenotopic organisms did not have enough time to develop a dominance over the eurytopic organisms. Deeper water conditions and increased 1975 SHAAK: BRUSH CREEK 113 substrate stability provided suitable habitats for the stenotopic organism blooms, thus pumping up the diversity of the fauna but without exclu- sion of the eurytopic components. It is quite possible for an explosive opportunist to numerically peak within the stillstand or stable-mature fauna if the limiting factor of that phase is temporal brevity. Glabrocingulum peaks numerically within the stable-mature phase and tails off in the relict-mature phase to a level intermediate between the transgressive and stillstand phases ( Table 13). The intermediate re- gressive value further supports the contention that Glabrocingulum is eurytopic. Plagioglypta was a sublittoral benthic infaunal animal ( "collector"), living partly embedded in the substrate with only the posterior extremity projecting into the water ( Ludbrook 1960: 137). Because of its restric- tion to keep ( sublittoral) water, it is an excellent stillstand indicator. The frequency trend of Plagioglypta through the faunai succession is similar to that of Glabrocingulum, because its peak is within the still- stand and the regressive average is intermediate in value ( Table 13). HoweVer the transgressive frequency is increased by 100 percent in the stillstand, whereas the regressive frequency is reduced 50 percent from that of the stillstand. This pronounced stillstand frequency peaking is a further indication of the Stenotopic nature of this animal. There currently is marked controversy as to the exact biological amnity of Plagioglypta. The large forms ( + 10 or coarser) probably are true scaphopods. The smaller forms, many of which pass through a number 100 sieve, may represent polychaete worm tubes. However the characteristic concentrically wrinkled shell is discernible on all size specimens. Regardless of their biological affinity they are recognizable, and their frequency trends place them within the stillstand or stable- mature fauna. Further supportive of a stillstand community is the frequency trend of the ornamented ostracods. The frequency averages of the orna- mented ostracods peak within the stillstand, with the transgressive and regressive frequency ·averaging significantly lower ( Table 13), Addi- tional support is provided by crinoid frequencies, which also peak within the stillstand. The successional maturity of the Glabrocingulum-Plagioglypta com- munity is quite low, as evidenced by the abundance of eurytopic com- ponents. Increased water depth and near-normal marine conditions existed, even if only temporarily. The Glabrocingulum-Plagioglypta community is dominated by epifaunal deposit feeders of the "collector" group, as is the case of the transgressive Astartella-Ianthanopsis com- munity. The most notable changes are in the sessile benthonic suspen- 114 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 sion feeders. The crinoids, ramose and fenestrate bryozoans, and brachi- opods reach their frequency peaks within this community. Environ- mentally, this community is farthest from shore, in the deepest water ( with reference to the study area), and .in a low terrigenous influx re- gime as is required by the sessile benthonic suspension feeders ("Fil- terers" A and B). The substrate is more stable than in the transgressive environments, as evidenced by the marked increase in the rooted and cemented benthos. Aclisina (OMMUNITY The Aclisina community represents the regressive phase and the cor- responding relict-mal:ure fauna. Aclisina is a high-spired snail with distinct spiral ornamentation of the family Murchisoniidae. It was prob- ably eurytopic, as evidenced by the similar frequencies throughout the marine event. However the frequencies increase gradually through the transgressive and stillstand phases and peak within the regressive phase ( Table 13). Aclisina is selected as name bearer for the relict mature community because it is cosmopolitan to the study area and is the only taxon of the cosmopolitan taxa that has its frequency peak within the regressive phase. The Aclisina community is the highest stratigraphically ( thus young- est) of the three laterally persistent benthic communities within the bounds of the study area. Environmentally, the Aclisina community is interpreted as indicating a return nearer to shore and shoal water conditions. It is the relict- mature communty of the regressive phase, composed of the more eury- topic components left over from the Glabrocingulum-Plagioglypta still- stand community. The stenotopic components are biologically win- nowed from the fauna during the entirety of the regressive phase. Thus the environment shifts from a low level of biological accommodation to pronounced physical accommodation. The opportunistic community is the product of aggradation, whereas the relict mature community is the product of degradation. The stable-mature community was marked by deeper water, offshore environments, and a decrease in terrigenous influx that allowed the more stenotopic sessile suspension feeders ( cri- noids and bryozoans) to increase in density. This process is reversed in the regressive phase. The only noticeable differences between the transgressive and regressive phases are in the vagaries of the faunal suc- cession. The explosive opportunists such as Astartella and Ianthanopsis never regain their prominence in the regressive phase. 1975 SHAAK: BRUSH CREEK 115 Rhombopora-Septopora COMMUNIT'Y The cosmopolitan communities are easily recognized and can be readily documented. However, one laterally restricted unique faunal association occurs within the study area and warrants discussion. The use of community in this instance is a convenient abstraction. Although this restricted faunal assemblage is a real three dimensional unit, it is a small scale one that cannot be recognized on a basin-wide level. The marine transgressive phase at Loc, 4, Sewickley, Pennsylvania, yielded an exquisite bryozoan fauna composed of the genera Septopora and Rhombopora. Septopora is a very distinctive cryptostomatous bryozoan of the family Acanthocladiidae. The genus is recognized by its distinctive branch-bifurcation network and associated dissepiments. Also, each branch is divided by a medial carina separating the laterally adjacent zooecial rows. Rhombopora is a rhabdomesid cryptosome found as ramose zoarial fragments displaying irregularly spaced bifurca- tions. The most striking structures are large blunt hpines that are the surBcial expressions of megacanthopores ( Huffman 1970: 675). The Rhombopora-Septopora community developed within the trans- gressive or opportunistic phase, culminating in the limestone subjacent to the stillstand sample. This community represents the highest bryo- zoan density within the study area. The locality itself is classified in the low diversity molluscan group of localities characterized by offshore, deeper water environments. Macroscopically the community is domi- nated by bryozoans, which occur as thick mats exposed on shale and limestone bedding planes. However, microscopically the community is numerically dominated by chonetid brachiopods, agglutinated forams, and pleurotomariacean and bellerophontacean snails. The community perhaps developed in a bryozoan garden or biostrome; thus the actual find was by chance, as the community probably represents just an areal patch. The community formed offshore in relatively deep water. The bryozoans are classified as "Filterers B" in the Turpaeva classificatory scheme and require deep offshore turbid waters under conditions of low terrigenous influx on a moderately stable subtrate. The bedding plane occurrence of the bryozoans perhaps represents alternating periods of silting and quiescence. DIVERSITY INDICES AS PALEOECOLOGIC TOOLS GENERAL STATEMENT Paleoecological and ecological research in recent years has shown a trend toward quantification of data. This quantification is applicable 116 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 to animal communities because of their inherent feature of diversity. Because animal diversity can be quantified, critical community features such as structure, stability, and evolution can be synthesized. One of the more informative methods of quantification is the derivation of numerical species diversity. COMPARISON OF THE SHANNON-WIENER AND SIMPSON DIVERSITY FUNCTIONS The Shannon-Wiener information function is sensitive to both the number of taxa present and the degree of evenness by which the number of individuals are distributed among the taxa. Stable values are rapidly reached and maintained when a population size of about 200 individuals for high stress environments and about 400 individuals for low stress environments per sample is achieved ( Sanders 1968: 279). With the recovery methods used in this investigation, minimum frequencies in most samples have been greatly exceeded. Therefore within the frame- work of this investigation the Shannon-Wiener function is relatively sample-size independent. There is, however, a very strong influence re- sulting from the evenness of spread of the individials among the various taxa. The Simpson function is markedly influenced by sample size and less by the evenness of spread of the individuals among the various taxa. As more individuals are added to the samples, there is a corresponding increase in diversity values. This relationship stems from the fact that individuals are added at a constant arithmetic rate, whereas taxa are added at a decreasing logarithmic rate. Because these two diversity indices are influenced by several vari- ables, both were used for all samples in the calculation of diversity values. A comparison of the two diversity index curves for each locality ( Fig. 3 to 11) shows a remarkable closeness of fit or parallelism. The index curves at localities 2 and 4 ( Shelocta and Sewickley) are practically congruent. The remaining localities show a close parallelism between the diversity index peaks. The Simpson index values are nu- merically higher than the Shannon-Wiener values, and the diversity value highs for the Simpson function are markedly more peaked than the corresponding Shannon-Wiener values. This pumping up of the Simpson peaks oyer the Shannon-Wiener peaks leads to the conclusion that these particular samples represent an increase in the number of taxa over those from the subjacent samples. This is not always true, however. Because of the individual frequencies evenness of spread in several sample analyses, the Simpson high peaks are controlled by high equi- tability values. As expected, the Shannon-Wiener peaks, except under 1975 SHAAK: BRUSH CREEK 117 conditions of low sample frequencies, are controlled by the equitability factor. The number of taxa retrieved with corresponding high numerical frequencies in this investigation generally average out the differences between the Simpson and Shannon-Wiener functions. Even though in- flections in the curves for the Simpson function are more pronounced than in Shannon-Wiener, the Simpson index values and curves, if used alone, are less reliable. This unreliability results from the dual influence of both equitability and taxonomic frequencies. RESTRICTIONS IN THE USE OF DIVERSITY INDICES A review of the data plots ( Figs. 3 to 11), including the diversity index, number of taxa, and equitability curves, supports the contention that the diversity index values and curves cannot safely be used alone to define communities or their trends. The diversity index curves range from peaked to flat, and the general shape of the curves can be con- trolled by masked subtleties in the faunal distribution. Two very real situations that are difficult to detect are patchiness, as previously de- scribed, and faunal aggregations. Faunal aggregations are a function of sampling if these aggregates are not due to taphonomy. Two distinct vertical faunal assemblages might be grouped within a single sample. An ectone might be sampled, showing combined effects of two adjacent communities. Unfortunately these vagaries generally cannot be detected in the outcrop. A faunal aggregation or ecotone commonly will cause an inflection on the diversity index curves that could go unexplained without further investigation. Shallow benthic marine communities, in general, are controlled by the density and diversity of the molluscs. Molluscan densities and di- versities increase towards shore, whereas the remaining faunal elements increase offshore ( Stevens 1971: 406). Thus the relationship of the molluscan species diversity curves to non-molluscs and total fauna quite often are as informative, or more informative, thanthe diversity index curves. Inflections on some diversity index curves.can be interpreted by molluscan faurial changes from the subjacent sample. A third tool, in addition to diversity index and number of taxa, is equitability. Some of the anomalous inflections on the diversity index curves, especially the Shannon-Wiener index curves, can be interpreted by inflections on the equitability curves. Thus a careful study of the relationship among diversity, index, species diversity, and equitabilty curves can provide a relatively sound interpretation of faunal succession and community structure. If the 118 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 diversity index curves were smooth, with the only inflection due to still- stand, sea level history would be easy to interpret. However the many factors that influence faunal distributions make this type of investigation very rewarding, because these factors can be reasonably interpreted by these combined techniques. CONCLUSIONS The classical interpretation of marine cycles places the stillstand and accompanying maximum diversity faunas at the position of the limestone unit. However this investigation shows that the limestone unit generally represents a transgressive stage. At only tw6 places, within the limits of the study area does it occur at the stillstand position. Also, there is a marked asymmetrical faunal distribution above and below the still- stand. Faunal successions, recurrent fossils, and suites of f6ssils were used to reconstruct laterally persistent fossil communities. The transgressive phase includes the Astartella-Ianthanopsis community, the stillstand phase supports the Glabrocingulum-Plagioglypta community, and the regressive phase includes the Aclisina community. Communities are named for the dominant genus or genera that occur in each. The structure of each community is inferred from recent ma- rine benthic communities of similar substrates, as well as presumed habitat and feeding preferences of the extinct animals. One localized exotic community, the Rhombopora-Septopora com- munity, is interpreted as a faunal garden or biostrome within the trans- gressive phase at one of the deep-water localities. This investigation also shows that the use of a diversity index as the primary interpretive tool in a paleoecological study can be misleading. A combination of diversity indices, equitability, and faunal composition provides an excellent system for a relatively sound interpretation of not only faunal succession but also community structure. This investigation shows that the limestones with a high terrigenous content generally are more Mable than the low terrigenous limestones. Limestone terrigenous content also appears to be one of the limiting factors in the level of biological accommodation reached by the faunas ~ recovered from limestone samples. One of the more significant results of this investigation is the mollusc, non-mollusc relationship. Molluscan faunas increase in both density and diversity nearshore, whereas the non-molluscan faunas increase offshore. Thus, not only the faunal succession but also the relative water depths for each locality throughout the marine event can be inferred from mollusc, non-mollusc trends. 1975 SHAAK: BRUSH CREEK 119 LITERATURE CITED Ager, D. V. 1963. Principles 6f paleoecology. New York, McGraw-Hill. 371 pA Beerbower, J. and D. Jordan. 1969. Application of information theory to paleontol- ogic problems: taxonomic diversity. J. Paleon. 43: 1184-1198. Benson, R. H. 1961, Ecology of ostracod assemblages. pp. Q56-63. In Moore, R. C., ed.; Treatise on invertebrate paleontology, pt Q, Arthropods. 3: 442 pp. Boardman, R. S. and A. H. Cheetham. 1969. Skeletal growth, intracolony varia- tion, and evolution in Bryozoa: a review. J- Paleon. 43: 205-233. Branson, C. C. 1962. Pennsylvanian System of central Appalachians. pp. 97-116. In Branson, C. C., ed. Pennsylvanian System in the United States. Am. Assoc. Petroleum Geologists Symposium. 508 pp. Brant, L. A. 1971. A study of the faunal succession in the Brush Creek Shale ( Pennsylvanian) near Shelocta, Pennsylvania. Unpublished M.S. thesis, Penn State Univ. Bretsky, F. W. 1969. Evalution of Paleozoic benthic marine communities. Palaeogeog., Palaeoclimatol., Paleoecol., 6: 45-59. Burke, J. J. 1930. The fauna of the Ames Limestone from Painter Hollow, Wells- burg, West Virginia. Unpublished M.S. thesis, Univ. of Pittsburgh. Buzas, M. A. and T. G. Gibson. 1969. Species diversity: benthonic foraminifera in western North Atlantic. Science, 163: 72-75. Comfort, A. 1964. Ageing, the biology of senescence. New York, Holt, Rinehart and Winston, Inc. 365 pp. Donahue, J. and M. Carothers. 1972. Diversity indices in paleoecology. Abst., Geol. Soc. Amer. Ann. Mtg., Minneapolis: 489. Donahue, J. and H. B. Rollins. ( in press). Paleoecological anatomy of a Cone- maugh ( Pennsylvanian) marine event. In Briggs, G., ed.; Carboniferous of the Southeastern United States. Geol. Soc. Amer., Spec. Paper No. 148. Donahue, J., H. B. Rollins, and G. D. Shaak. 1972. Asymmetrical community sue- cession in a transgressive-regressive sequence. 21st Int. Geol. Congress: 74-81. Doney, H. H. 1954. Pelecypods of the COnemaugh Series from Pittsburgh, Penn- sylvania and neighboring areas. Unpublished M.S. thesis, Univ. of Pittsburgh. Dunbar, M. J. 1960. The evolution of stability in marine environments. Natural selection at the level of the ecosystem. Amer. Nat. 94: 129-136. Emery., K. O. 1950. Ironstone concretions and beach ridges of San Diego County, California. Calif. J. Mines and Geol. 46: 213-221. Fagerstrom, J. A. 1964. Fossil communities in paleoecology. Geol. Soc. Amen Bull. 75: 1197-1216. Feder, H. and A. M. Christensen. 1966. Aspects of asteroid biology. pp. 87-128. In Boolootian, R. A., ed.; Physiology of Echinodermata. New York, Interscience Publishers. 822 pp. Fell, H. B. 1966. The ecology of ophiuroids. pp. 129-144. In Boolootian, R. A., ed.; Physiology of Echinodermata. New York, Interscience Publishers. 822 pp. - and D. L. Pawson. 1966. Ecology of crinoids. pp. 49-62. In Boolootian, R. A., ed.; Physiology of Echinodermata. New York, Interscience Publishers. 822 pp. Flint, N. K. 1965. Geology and mineral resources.of southern Somerset County. Pa. Geol. Surv. Bull. (56A: 267 pp Griesemer, A. D, 1970. Paleoecology of the Ervine Creek Limestone ( Late Penn- sylvanian) in the miscontinent region. Unpublished Ph.D. thesis, Univ. Ne- braska. Grimths, J. C. 1967. Scientific method in analysis of sediments. New York, McCraw-Hill. 508 pp Hoskins, D. M. 1969. Fossil collecting in Pennsylvania. Pa. Geol. Surv. Bull. 640: 126 pp 120 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 Huffman, S. F. 1970. The ectoproct ( Bryozoan) Rhombopora lepidodendroides Meek, Late Pannsylvanian ( Virgilian), Nebraska. Paleon. 44: 673-679. Hymen, L. 1959. The invertebrates: smaller coelomate groups. New York, McGraw-Hill. 783.pA Johnson, R. G. 1962. , Interspecific associations in Pennsylvanian fossil assemblages. Geol. 70:32-55. - 1964. The community approach to paleoecology. pp. 107-134. In Imbrie, J. and N. D. Newell, eds.; Approaches to paleoecology, New York, John Wiley and Sons, Inc. 432 pp. - 1970. Variations on diversity within benthic marine communities. Amer. Nat. 104: 285-300. - 1972. Conceptual models of benthic marine communities. pp. 148-159. In Schopf, T.J.M., ed.; Models in paleobiology. San Francisco, Freeman Cooper and Co. 250 pp. Kesling, R.. V. 1961. Ontogeny of Ostracoda. pp. .Q19-20. In Moore, R. C., ed.; Treatise on invertebrate paleontology, pt. Q. Arthropoda 3: 442 pp. Leighton, H. 1947. Guidebook to the geology about Pittsburgh. Pa. Geol. Surv. Bull. 617: 35 pp Levinton, J. S. . 1970. The paleoecological significance of opportunistic species. Lethaia. 3: 69-78. Lintz, J., Jr. 1958. The fauna of the Ames and Brush Creek Shales of the Cone- maugh Formation of western Maryland. J. Paleon. 32: 97-112. Lloyd, M. and R. J. Ghelardi. 1964. A table for calculating the "equitability" component of species diversity. Animal Ecol. 33: 217-225. Ludbrook, N. H, 1960. Scaphopoda. pp, 137-41. In Moore, R. C., ed.; Treatise on invertebrate paleontology, pt. 1, Mollusca. 1: 351 pp. Lund, R. 1970. A new technique for chemical preparation of fossils. J. Paleon. 44: 578. Margalef, R. 1963. Successions of populations. Adv. Frontiers Plant Sci. 2: 137- 188. - - 1968. Perspectives in ecological theory, Univ. Chicago Press. 111 pp. Mark, C. G. 1912. The fossils of the Conemaugh Formation in Ohio. Ohio Geol. Surv., 4th Ser., Bull. 17: 261-318. Miller, A. R. and .A. G. Unklesbay. 1942. The cephalopod fauna of the Conemaugh Series in western Pennsylvania. Carnegie Mus, Annals. 29: 127-174. Moore, H. B. 1966. Ecology of echinoids. pp. 73-83. In Boolootian, R. A. ed.; Physiology of Echinoderinata. New York, Interscience Publishers. 822 pp. Moore, R. C. 1944: Correlation of Pennsylvanian Formations of North America. Geol. Soc. Amen Bull. 55 657-706. Morris, D. A. 1967. Lower COnemaugh ( Pennsylvanian) depositional environ- ments and paleogeography in the Appalachian coal basin. Unpublished Ph.D. thesis, Univ. of Kansas. Murphy, J. L. 1967. R. P. Steven's pelecypod species from the Brush Creek shale member ( Conemaugh) of Ohio. J. Paleon. 41: 1498-1504. - 1970. C6iled nautiloid ce-phalopods from the Brush Creek Limestone ( Conemaugh) of eastern Ohio and western Pennsylvania. J. Paleon. 44: 195- 204. Newell, N. D., et at. 1959. Organism communities and bottom facies, Great Ba- hama Bank. Amer. Mus. Nat. Hist. Bull. 117: 117-228. Odum, E. P. 1971. Fundamentals of ecology. Philadelphia, W. B. Sanders Co. 574 pp. Pawson, D. L. 1966. Ecology of holothurians. pp. 63-72. In Boolootian, R. A., ed.; Physiology of Echinodermata. New York, Interscience Publishers. 822 pp. Petersen, C. G. J. 1911. Valuation of the sea. I, animal life of the sea-bottom, its food and quantity. Rept. Danish Biol. Sta. 20: 81 pp. 19I3. Valuation of the sea. II, the animal communities of the sea-b6ttom 1975 SHAAK: BRUSH CREEK 121 and their importance for marine zoogeography. Rept. Danish Biol. Sta. 2: 44 pp - 1914. Appendix to report 21. On the distribution of the animal com- munities of the sea-bottom. Rept, Danish Biol. Sta. 22: 7 pp. - 1915. On the bottom communities.of the sea-bottom in the Skagerrak, the Christina Fjord, and the Danish waters. Rept. Danish Biol. Sta. 23: 3-8. - 1918. The sea-bottom and its production of fish-food. A survey of the work done in connection with the valuation of the Danish waters from 1883- 1917. Rept. Danish Biol. Sta. 25: 62 pp. Platt, W. G. 1879, Report of progress in Armstrong County. Pa. End Geol. Surv. Rept. HHHH: 338 pp. Purdy, E. G. 1964. Sediments as substrates; pp. 238-271. In Imbrie, J. and N. D. Newell, eds.; Approaches to Paleoecology. New York, John Wiley and Sons, Inc. 432 pp. Raymond, P. E. 1910. A preliminary list of the fauna of the Allegheny and Cone- maugh Series in western Pennsylvania. Carnegie Mus. Annals. 7: 144-158. Rogers, H. D. 1858. The geology of Pennsylvania. Pa, Ist Geol, Surv. 1. 586 pp, 2: 815 pp. Rollins, H. B. and J. Donahue, 1971. Faunal community asymmetry in the Cone- maugh ( Upper Pennsylvania) of the Appalachian Basin: theoretical implications. Abst., Geol. Soc. Amen Ann. Mtg Washington, D. C.: 687. - and · -. 1972. Transgression and regression: relative rates based on faunal communities. Am. Assoc. Petroleum Geologists, Eastern Section, 1st Ann. Meeting, Columbus, Ohio: 67-81. - and -. 1972. Paleoecological applications of the stability-time hypoth- esis. Abst., Geol. Soc. Amer. Ann. Mtg., Minneapolis: 643. Rudwick, M. J. S. 1970. Living and fossil brachiopods. London, Hutchinson University Library. 199 pp. Ryland, J. S. 1970. Bryozoans. London, Hutchinson University Library. 175 pp. Sanders, H. L. 1968. Marine benthic diversity: a comparative study. Amer. Nat. 102: 243-282. ----. 1969. Benthic marine diversity and the stability-time hypothesis. pp. 7i-81. In Diversity and stability in ecological systems. Brookhaven Symposia in Bi- ology. Nd. 22: 264 pp. Seaman, D. M. 1940, The Ames Limestone of western Pennsylvania. Pa. Acad. Sci. Proc. 14: 77-80. - 1941. The Cambridge ( Pine Creek) Limestone of' western Pennsylvania. Pa. Acad. Sci. Proc. 15: 60-65. 1942. The Brush Creek Limestone of western Pennsylvania. Pa. Acad. Sci. Proc. 16: 72-76. Shaak, G. D. and H. B. Rollins. 1972. Reflection of community standing crop in Pennsylvanian faunal assemblages. Abst., Geol. Soc. Amer. Ann. Mtg., Min- neapolis: 661. Shabica, C. W. 1970. Depositional environments in the Francis Creek Shale. pp. 43-52. In Smith, W. H., et al., eds:, Depositional environments in parts of the Carbondale Formation-western and northern Iliinois. Ill. State · Geol. Surv. Guidebook Ser. No. 8: 119 pp. Simpson, E. H. 1949, Measurement of diversity. Nature. 163: 688. Speden, I. B. 1966. Paleoecology and the study of fossil benthic assemblages and communities. New Zealand J. Geol. and Geog. 9: 408-423. Stevens, C, H. 1965. Faunal trends in near shore Pennsylvanian deposits near McCoy, Colorado. Mountain Geologist. 2: 71-77. - 1966. Paleoecological implications of Early Permian fossil communities in eastern Nevada and western Utah. Geol. Soc. Amer. Bull. 77: 1121-1130. - 1969. Water depth control of fusulinid distribution. Lethaia. 2: 121-132. 122 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 - 1971. Distribution and diversity of Pennsylvanian marine faunas relative to water depth and distance from shore. Lethaia. 4: 403-412. Sturgeon, M. T. and R. D. Hoare. 1968. Pennsylvanian brachiopods of Ohio. Ohio Geol. Sum Bull. No. 63: 95 pp. Theiss, M. E. 1940. The fauna of the Ames Limestone in the Pittsburgh Quad- rangle. Unpublished M. S. thesis, Univ. of Pittsburgh. Thorson, G. 1955. Modern aspects of marine level-bottom animal communities. J. Marine Research. 14: 387-397. -. 1957. Bottom communities ( sublittoral or shallow shelf). pp. 461-534. In Hedgpeth, J. W., ed.; Treatise on marine biology and paleoecology, v. 1. Geol. Soc. Amer. Mem. 67: 1296 PA -. 1971. Some factors infuencing the recruitment and establishment of marine benthic communities. pp. 147-173. In Nybakken, J. W., ed.; Readings in marine ecology. New York, Harper and Row. 544 pp. Turpaeva, E. P. 1957. Food interrelationships of dominant species in marine benthic biocoenoes. Trudy Inst, Okeanol. Akad. Nauk. SSSR 20: 171-185. Valentine, J. W. 1971. Resource supply and species diversity patterns. Lethaia. 4: 51-61. Wagner, W. R., et at. 1970. Geology of the Pittsburgh area. Pa. Geol. Surv. Gen. Geology Rept 659: 145 pp. West, R. R. 1970. Marine communities of a portion of the Wewoka Formation ( Pennsylvanian) in Hughes County, Oklahoma. Unpublished Ph.D., thesis, Univ. of Oklahoma. White, I. C. 1878. Report of progress in the Beaver River District of the bi- tuminous coal fields of western Pennsylvania. Pa. 2nd Geol. Surv. Prog. Rept. Q: 335 pp. Whittaker, R. H. 1965. Dominance and diversity in land plant communities.Science. 147: 250-260. Wigley, R. L. and A. D. Mcintyre. 1964. Some quantitative comparisons of off- shore meiobenthos and macrobenthos south of Martha's Vineyard. Limologyand Oceanography. 9: 485-493. Williams, E. G. 1960, Marine and fresh water fossiliferous beds in the Pottsvilleand Allegheny Groups of western Pennsylvania, J. Paleon. 34: 908-922.Williams, H. S. 1891. Correlation papers: Devonian and Carboniferous. U. S.Geol. Surv. Bull. 80: 297 pp. Wilson, E. 0. 1969. The species equilibrium. pp. 38-47. In Diversity and stabilityin ecological systems. Brookhaven Symposia in Biology. No. 22: 264 pp.Wilson, J. L. 1967. Cyclic and reciprocal sedimentation in Virgilian strata ofsouthern New Mexico. pp. 82-99. In Cyclic Sedimentation in the PermianBasin. West Texas Geol. Soc. Symposium, 219 pp. Zangerl, R. and E. S. Richardson, Jr. 1963. The paleoecological history of twoPennsylvanian black shales. Fieldiana: Geol. Mem., Chicago Nat. Hist. Mus.4. 352 pp 1975 SHAAK: BRUSH CREEK 123 APPENDIX A.-MEASURED SECTIONS SHOWING SAMPLING INTERVAL AT SAMPLE STTESI eet KITTANNING LOCALITY - 55 -- 50 45 - shale 40 35 -.. sil*stone 30 -- . !imestone coquina 25 1 0 6-c. 4 (- nodular silfs,one sample lilli ; limestone interval I. 20 bedded nodules 15 sillstone 10 shole 5 -0 coal LOCALITY 1.-Stratigraphic section and sample interval at Kittaning, located along northside of U.S. 422, 0.7 mi. E. of intersection of U.S. 422 and Pa. 663, Arm- strong County ( Cadet Drive-In). 1 All localities are in Pennsylvania. 124 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 eet SHELOCTA LOCALITY - 30 -r- mudstone TT 25 - . C. -- 9 0 20 0 C nodular shale -- 0 0 <3 0 15 ° 00= -- -- 0 -- sample ,O 11~'1'1'1' limestone interval - ~ nodular shale --0 1 ./ T/.T nodular mudstone 5 -T--7- shale -0 coal LOCALrry 2.-Stratigraphic section and sample interval at Shelocta, located along northside of U.S. 422, 0.5 mi. W. of intersection of U.S. 422 and Pa. 156, Indiana County ( adapted from Brant 1971). 1975 SHAAK: BRUSH CREEK 125 feet BARTON LOCALITY - 25 -20 -% interbedded sandstone and shale shale - 15 - 10 - ==== calcareous shale '=:=DI--1 - 1--1-1 nodular Ilmes'one sample - Illili/ interval ==== calcoreous shale 5 shale railroad bed LOCALITY 3.-Stratigraphic section and sample interval at Barton, located along Baltimore and Ohio Railroad cut on eastside of U.S. 119, 1.9 mi. N. of intersec- tion of U.S. 119 and Pa. 403 ( Marion Center, Pa.), Indiana County. 126 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 Pet SEWICKLEY LOCALITY -10 I shc,le -.. nodular sillsione - •• ~ sillstone Ill sample ~ | ' limestone interval 1 1 5 111 shale -0 LocALrrY 4.-Stratigraphic section and sample interval near Sewickley, located at a roadcut along southside of Pa. 51, at south end of Sewickley Bridge, Alle- gheny County. 1975 SHAAK: BRUSH CREEK _ 127 pet GLENSHAW LOCALITY -20 shale -15 - - mudstone sample - „=c==\ nodular mudstone interval - 10 | limestone - T -r - - -T- - -1- massive mudstone -5 - -r - -r -T -r - -r -r -0 LocALrry 5.-Stratigraphic section and sample interval at Glenshaw, located at an excavation along westside of Pa. 8, at intersection of Pa. 8 and Maple Street, Allegheny County ( Glenshaw Glass Company). 128 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 feet MURRYSVILLE LOCALITY- - 10 0 nodular siltstone 0(= -.. -5 5 0 - 0 00.... bedded nodules -- shale sample interval ~|~|~ limestone (channel) sillstone ~ 1 ~| ~ limeNone -[-~- ( carbonaceous shale ~ coal LocALrry 6.-Stratigraphic section and sample interval at Murrysville, located at an excavatibn on northside of U.S. 22, to immediate east of Allegheny-Westmore- land County line, Westmoreland County ( William Penn Lumber Company). 1975 SHAAK: BRUSH CREEK 129 feet PICCOLOMINI LOCALITY -60 -T--1- -r-r - 55 7- IT - 50 -r -T- -r -r T T mudstone -45 -T- -T - 40 T -r 7-T T -T - 35 - 30 h.:.·.1 siltstone und quartz wackestone -25 - 2--1 - 20 -.. -15 0 „- ~ nodular shole sample ~ ~,> 1 Ill limestone and ironstone interval -10 5 shale coal-0 LOCALITY 7.-Stratigraphic section and sample interval near Uniontown, located at abandoned Piccolomini Brothers strip mine, on eastside of Pa. 51, 5 mi. N. of Uniontown, Fayette County. 130 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 eet URSINA LOCALITY -30 · ) sands*one 25 20 _~ shole 15 '0 l limestone sample 00 interval -- - - nodular shale 01 5 c- ~ i shale -0 LOCALITY 8.-Stratigraphic section and sample interval near Ursina, located along westside of Jersey Church Road, 0.95 mi. N.W. of intersection of Pa. 53 and Jersey Church Road, Somerset County (adapted from Flint 1965). 1975 SHAAK: BRUSH CREEK 131 feet GLADE CITY LOCALITY-35 "*j sandstone 30 25 siltstone 20 la ~- shale 10 Iii,11 channel limestone0 -2 -- a nodulor shale sample 1 1 1-1 1 1 limestone interval - --- --_ shole -..- nodular sillstone 5 shale coal -0 LOCALrry 9.-Stratigraphic section and sample interval at Glade City, located along northside of Western Maryland Railroad, 1.5 mi. E. of Meyersdale, Somerset County ( adapted from Flint 1965). 132 BULLETIN FLORIDA STATE MUSEUM Vol. 19, No. 2 APPENDIX B.-DATA MATRIX OF BRUSH CREEK FosSILS. TAXON LOCALITY 12345 789 Foraminifera Cyclogyra sp XXXXX X X 0 X X X X X X Tolipammina sp. XXXX X X Ammobaculites sp. XXXX X X Ammovertella sp. X X X Glomospira sp. X X X X X Thurammina sp. X Lituotuba sp. X X X X X X Ammodisous sp. X X Paleotextularia sp. XXX Psammosphaera sp. X Reophax sp. XXX Hemigordius sp. X Bigenerina sp Huperammina sp. X X X ) X X X X X lagenid Ostracods Healdia-Bairdia sp. X X X XXX ( smooth) Amophissitids ( pitted) XXXX XXX X X X X Hollinella sp. C nodose ) X X X XXX Porifera Spicules X Coelenterata Stereostytus sp. XXX X X Brachiopoda Crurithyris sp. X X X X X Juresania sp. X X Neochonetes sp. X X X X X Linoproductus sp. X Composita sp. 1 X X X > < > 1 Derbya sp. Antiquatonia sp. X Bryozoa Ramose XXX XXXX Fenestrate X X Septopora biserialis X X Rhombopora tepidodendroides X Bivalvia Astartella sp. XXXXX XXXX Girtuana sp. X X X Nucula sp. X X X Culunana sp. X X X X Altorisma sp. X XXXX Nucutana sp. X X X X Naiadites sp. X Aviculopecten sp. X X X X X Septimt/alina sp X Anthraconaia sp. X Gastropoda Gtabrocingulum sp. XXXXXXXXX Pseudozygoplieurids X X X X ·- X X X 1975 SHAAK: BRUSH CREEK 133 APPENDIX B.-CONTINUED TAXON LOCALITY 12345678 Ianthanopsis sp. XXXXXXXX 63 X X X X X > < X X Meekospira sp. XX X Belleron sp. X Pharkidonatus percarinatus X X Euphemites sp. XXXXX Retispira sp. X X XXX Bulimorpha sp. X X X X X X >1 ,> 4 >1 X X X >