Paul Copper

Oxygen and carbon isotopic composition of Silurian brachiopods: Implications for coeval seawater and glaciations

We collected 236 calcitic brachiopod shells, covering the entire Silurian Period (~ 30 m.y.), at high temporal resolution from stratotype sections from Anticosti Island (Canada), Wales (United Kingdom), the Oslo region (Norway), Gotland (Sweden), and Podolia (Ukraine), Estonia, Latvia, and Lithuania. Data from petrography, scanning electron microscopy, cathodoluminescence, isotopes, and trace elements all confirmed that there was excellent preservation in most shells, thus arguing for retention of primary isotope signals; exceptions were samples from the Oslo region. The d18O and d13C values for the well-preserved samples range from –2‰ to –6.5‰ and from –1‰ to 7.5‰ (Peedee belemnite), respectively. In terms of temporal trends, oxygen and carbon isotopes vary in parallel, with a slight decrease with declining age of ~1‰ through Silurian time, with superimposed short-term oscillations that are negatively correlated with sea-level changes. Three successive positive d18O shifts in early Aeronian, latest Aeronian, and early Wenlock time correlate with sea-level lowstands and with glacial diamictite deposits in the Amazon Basin and in Africa. The high d18O and d13C values are attributed to cold episodes with low sea levels and low values to warm episodes with high sea levels. During warm Silurian episodes, the d18O value of seawater is suggested to have been about –3.5‰ standard mean ocean water (SMOW) and the global tropical temperatures of about 20–30 °C, similar to the present-day values in summer. During glacial episodes, seawater is proposed to have had a d18O value of ~–2.5‰ (SMOW) and temperature of about 14–23 °C, comparable to tropical temperatures proposed for the last glacial episode (14 to 28 °C). Three positive d13C peaks, in early and late Wenlock and late Ludlow time, likely of regional to global significance, appear to coincide with sea-level lowstands, but we are as yet unable to propose a convincing causative geologic scenario that would explain their origin.

Frasnian/Famennian mass extinction and cold-water oceans

Suturing of Laurussia and Africa, elimination of easterly tropical currents, and consequent equatorial diversion of cold and possibly dysaerobic waters are suggested as prime causes of the Frasnian/Famennian (Late Devonian) extinction. These are explained by using a new paleogeographic model. Evidence for stepdown extinctions that took place from Givetian through Frasnian time is taken particularly from atrypoid brachiopods, one of the most abundant representatives of the global tropical shelly marine benthos during the Middle and Late Devonian. Only 5 of 17 Frasnian genera or subgenera of atrypoids are known to be present in late Frasnian sediments, and possibly only a single genus, Spinatrypa , was present at the Frasnian/Famennian boundary. Evidence for stepdown extinction from other invertebrate and vertebrate groups in general corroborates that provided by brachiopods.

Reef development at the Frasnian/Famennian mass extinction boundary

A newly compiled global reef database indicates that the 5^6 Myr long Frasnian (Late Devonian) metazoan reef episode had relatively lowdiversity compared to Middle Devonian highs (w ith over 200 genera of calcitic rugose and tabulate corals). Following an initial early rise after Late Givetian coral and stromatoporoid extinctions, reefs expanded for the last time during mid-Frasnian sealevel highstands, but declined markedly in the Late Frasnian (rhenana-linguiformis conodont zones), belowthe Frasnian/Famennian (F/F) boundary. Globally, metazoan reefs were wiped out by the end Frasnian: some Famennian reefs, while partly retaining the structure of the underlying carbonate platform, were built by cyanobacterial consortia such as Renalcis, Rothpletzella, Girvanella and Epiphyton. During the Famennian, foraminiferans with calcite walls became abundant for the first time in the Phanerozoic, adding a newdimension to carbonate platforms. Colonial rugose corals (phaceloid, cerioid and thamnasterioid modules) were absent in the early post-extinction phases up into the mid-Famennian, and very rare and non-reefbuilding later, but solitary deep-water Lazarus corals survived locally. Coral^sponge reefs are unknown from the 21 Myr long Famennian, also a time of very lowplatform carbonate production. Rare, small, isolated stromatoporoid sponge, and lithistid sponge patch reefs returned episodically during the Famennian in North America, western Europe, Australia and China: the aragonitic stromatoporoids became extinct at the end of the Famennian. During a Late Devonian tectonically very active, collisional Caledonian mountain-building phase, oceanic and atmospheric cooling, accompanied by sealevel lowstand systems, exposed most carbonate platforms, accelerating coastal erosion and karsting. This increased the amount of clastics in the shelf-slope setting, in the last 1^3 Myr prior to the F/F boundary, often burying reefs. Immediately following, there were protracted losses in nearly all major tropical shelf, benthic marine invertebrates, exceeded only by the end Permian extinctions in severity. There is no apparent link between black, organic-rich horizons and reef demise at or close to the F/F boundary. The F/F boundary not also marks the largest change from widespread flooded Early and Mid-Paleozoic continental cratons to narrow, distal shelves, but also spikes the largest known global Phanerozoic shift in atmospheric O2 enrichment, and CO2 drawdown. This threshold matched the rise of the first tropical rainforests, and expansion of terrestrial biomes on the tropical coastal lowlands formerly occupied by carbonate platforms. 6 2002 Elsevier Science B.V. All rights reserved.

Ecological succession in Phanerozoic reef ecosystems; is it real?

I define ecological succession as an orderly, directional, and predictable, pioneer-to-climax process of community and species development. It takes place in an environment where external physico-chemical constraints are not undergoing major change: ecological succession leads naturally to increasing biological control of the environment and a stabilized ecosystem. Ecological succession is an intrinsic feature of reef growth. I present criteria for the recognition of pioneering and climax stages in Phanerozoic reefs at both the species and community level. On the other hand, I use the term community replacement where changing external factors have forced the migration of organisms into an area thereby replacing existing species and communities. Development of reef ecosystems through the Phanerozoic generally follows the pattern of ecological succession but through a time frame of 30 to 10OMa, concluded by a phase of mass extinction. I call such long-term changes erathemic successions. Pioneering recovery of reef ecosystems following extinctions appears to take as much as 8 to 12Ma. Restricted marine environments or global periods of climatic change produce arrested successions-successions essentially halted in initial phases characterized by pioneering species and communities that fail to reach a full climax stage. I outline selective examples of arrested successions. Arrested succession may explain the lack of full reef development within most of the mid-Cambrian to midOrdovician and much of the Carboniferous and may provide a key to explaining global climatic change. Classification of reefs should take into account the role of community structure and ecological succession as these may play a stronger selective role on reef type than tectonic, sedimentological, or physico-chemical constraints.

Paleolatitudes in the Devonian of Brazil and the Frasnian-Famennian mass extinction

Abstract Differences in faunal composition of contemporary Devonian marine communities in Brazilian sedimentary basins suggest variation because of climatic gradients. Temperatures may have ranged from subarctic conditions in the Parana Basin, a type of Devonian Hudsons Bay, to temperate in the Amazon and adjoining basins, analogous to a modern “north Atlantic” climate. These gradients were paralleled in other parts of South America. The Malvinokaffric cold climate fauna was dominated by groups oforganisms that survived the worldwide late Devonian mass extinctions. Invertebrate groups absent in the Malvinokaffric regions of Brazil, Bolivia and Argentina (i.e. those restricted to Devonian equatorial belts of North America, Eurasia and Australia) were decimated in the late Devonian. These two parallel observations suggest that late Devonian extinction may have been the result of drastic cold spells that killed off reefal and peri-reefal life. By elimination of crucial benthic community components, the trophic structure of the shallow marine ecosystem of the time was upset. Subsequent repopulation of the Carboniferous seas was accomplished by hardy, eurythermal invertebrate taxa present in cold as well as tropical regions. Cold spells and mass mortality in very shallow waters may go some way to explain the production of the widespread black shale environments so typical of marine regressive phases in the late Devonian of the western hemisphere.

Silurian strontium isotope stratigraphy

A sample set of 164 calcitic brachiopod shells, covering the entire Silurian Period (∼ 30 m.y.) with a resolution of about 0.7 m.y., was collected from stratotype sections at Anticosti Island (Canada), Wales (United Kingdom), Gotland (Sweden), Podolia (Ukraine), Latvia, and Lithuania. They show 87Sr/86Sr values ranging from 0.707930 to 0.708792 that progressively increase with time. This may indicate an increasing riverine flux of radiogenic Sr into the ocean from weathering of continental sialic rocks due to progressive warming of the climate. Exceptionally high increases in 87Sr/86Sr values were observed in early Llandovery (Rhuddanian), late Llandovery (Telychian), and late Ludlow (Gorstian-Ludfordian boundary) samples. Partial linear regressions, based on a stepwise climbing pattern, with local drops around the Llandovery-Wenlock boundary and in latest Ludlow time, were used to estimate relative ages with a resolution of about ±2 biozones (∼1.5–2 m.y.). The Sr-isotope curve shows distinct inflection points in earliest Wenlock and mid-Pridoli time. These may be used to correlate the Llandovery-Wenlock boundary in the United Kingdom, Gotland, and Lithuania, and the Kaugatuma-Ohesaare boundary in the Baltic states and Podolia.

Environmental and Substrate Control on Paleozoic Bioerosion in Corals and Stromatoporoids, Anticosti Island, Eastern Canada

Abstract Bioerosion was a common process affecting corals and stromatoporoids in reef and off-reef facies on the carbonate ramp that spanned the Ordovician–Silurian boundary on Anticosti Island. The probable worm boring Trypanites was the dominant macroboring, penetrating more than 40% of 2,500 massive tabulate corals and stromatoporoids examined, occasionally in dense concentrations. The frequency of macroboring was influenced by conditions at the facies level reflected by changes in grain size, water depth, storm reworking of sediments, and the nature of the skeletal mass bored. These factors regulated exposure time of the host-substrate surface to the watermass and thus influenced bioerosion. Bored specimens are most common in muddy off-reef facies, moderate in sandy off-reef facies, and less common in reefs. In off-reef facies, storm-enhanced deposition and reworking of sediments were most important in the burial of eligible host substrates. In reefs, the high competition for space by encrusting epizoans, combined with sedimentation, limited macroborers that preferred to excavate dead skeletal substrates. Skeletal density was the most important property of the host substrate in controlling boring frequency. Macroborers favored a dense host skeleton likely for its enhanced mechanical strength and adaptability for unlined borings, despite requiring greater energy for excavation. High-relief host skeletons were bored more frequently than tabular forms, since their greater capacity to shed sediment would have resulted in more prolonged exposure above the seafloor. The probable bivalve boring Petroxestes pera is rare. Temporal changes in boring frequency appear to reflect local shifts in facies and relative sea level. Mass-extinction events near the O/S boundary, which eliminated some host corals and stromatoporoids, had no apparent effect on boring frequency.

EPIPUNCTAE AND PHOSPHATIZED SETAE IN LATE ORDOVICIAN PLAESIOMYID BRACHIOPODS FROM ANTICOSTI ISLAND, EASTERN CANADA

Abstract Epipunctae, a new type of shell perforation, are well developed in typical taxa of the family Plaesiomyidae, a group of common orthide brachiopods from Laurentia and some other tropically located tectonic plates of Late Ordovician age. These minute, prominently elongate, tubular structures are similar to endopunctae in size and density, but differ in being oblique, intersecting the shell surface at a relatively low angle, and being confined largely to the outer portion of the shell wall. The tubules are similar in orientation to aditicules within the same shells but are much smaller and denser, usually aligned along fine growth lines and arranged in crude longitudinal columns. Exceptionally preserved phosphatic molds of bundled setal canals inside epipunctae and aditicules, described for the first time in this paper, are direct evidence that these two types of tubular structures of different sizes had the same function of housing sensory setae along the shell margin, but both the setae and the tubules became abandoned in the outer portion of the shell wall through burial by the secondary shell layer when the shell margin migrated forward. Epipunctae have been found so far only in plaesiomyid shells, but aditicules are common in many groups of the order Orthida. The taxonomic value of epipunctae is shown by a reassessment of Pionorthis Schuchert and Cooper, 1932. The hypotype previously regarded widely, but erroneously, as the archetype of ‘Orthis sola’ Billings, 1866, the type species of Pionorthis, is allied to Plaesiomys Hall and Clarke, 1892. It bears the characteristic epipunctae. The holotype of Orthis sola is a dalmanelloid shell with true punctae, assignable to Mendacella Cooper, 1930. This warrants rejection of the genus Pionorthis.

The deep-water brachiopod Dicoelosia King, 1850, from the Early Silurian tropical carbonate shelf of Anticosti Island, eastern Canada

Dicoelosia occurs in two deep water benthic shelly assemblages on an Early Silurian (uppermost Aeronian, Stimulograptus sedgwickii Zone) carbonate ramp to shelf, within the 25 m thick bluish-grey mudstone of the Richardson Member in the middle Jupiter Formation, Anticosti Island, Quebec. Dicoelosia dauphinensis new species is erected on the basis of its relatively large, elongate shell, with a concavo-convex lateral profile, moderately wide, planoconvex lobes and subparallel lateral margins. Dicoelosia dauphinensis first occurs in a Gotatrypa-Dicoelosia Community, in which it constitutes about 7 percent of the individuals within sampled populations. This is replaced about 2 m up section by a Resserella-Dicoelosia Community, where D. dauphinensis makes up 22 percent of the individuals preserved on single bedding planes. Subsequently, Dicoelosia becomes a rare component of the Stegerhynchus-Triplesia Community in a shallowing-upward succession of the upper Richardson Member. It is in this upper Richardson shelly community that Stimulograptus sedgwickii is locally common, together with in situ brachiopod nests of Eocoelia, Triplesia, and Lissatrypa at the upper Jupiter Cliff section. The water depth estimated for the Resserella-Dicoelosia Community is between 100-120 m, below the normal depth range of the Clorinda Community (BA5) on Anticosti, in a distal shelf setting about 80-100 km offshore from the Laurentia paleocontinent, on the west side of the Iapetus Ocean. The interpretation of water depth is based on the facts that, 1) Dicoelosia-rich communities are confined to strata, which lack shallow water sedimentological evidence such as thick calcarenites and hummocky cross stratification, 2) Dicoelosia does not occur with shallower water taxa seen in the overlying Clorinda and Stricklandia communities associated with cyclocrinitid algae, corals and stromatoporoids in the Cybele Member, and 3) Dicoelosia occurs only in the deeper water sections of the Richardson Member, some 10-15 km basinwards from mid-shelf shallower Richardson facies of the Anticosti Basin. Thus, the Dicoelosia-rich communities are interpreted to mark a maximum flooding surface within the distal shelf to ramp Llandovery succession of Anticosti Island.

Parastrophinella (Brachiopoda): Its paleogeographic significance at the Ordovician/Silurian boundary

The type species of Parastrophinella, P. reversa, a pentamerid brachiopod from the uppermost Ordovician (Hirnantian) Ellis Bay Formation of Anticosti Island, Quebec, shows that the genus is characterized by three features: 1) a ventral median septum apically buried in a thickened valve floor, but anteriorly rising above valve floor; 2) prominent alate plates, which are homologous to brachial processes; and, 3) a pseudocruralium consisting of a dorsal median septum, which is largely buried in the valve floor posteriorly, and outer plates that are connected to the median septum via prismatic substance (with poorly developed lamellar layer at the junctions) at, or slightly above, the valve floor. These constitute criteria by which many species previously assigned to Parastrophinella are excluded from the genus. Late Ordovician species that fit the redefinition of Parastrophinella are now confined to eastern North America. The genus crosses the Ordovician/Silurian boundary without major morphologic change, and Early Silurian species of Parastrophinella occur in both eastern North America and Great Britain. In the Late Ordovician, the pentamerid fauna of North America (Laurentia) contrasted sharply with that of Baltica and Kazakhstan, where the Holorhynchus fauna was dominant. Typical elements of the Holorhynchus fauna, such as the large-shelled Holorhynchus and Proconchidium, are known only in the northern parts of Laurentia (Baffin Island, Greenland, and Kolyma), and are absent in the Hudson Bay and Williston basins and southwards.