Abbreviations

AMNH, Division of Paleontology, American Museum of Natural History; GSI, Geological Survey of Israel, Jerusalem; NHM, Department of Palaeontology, Natural History Museum, London.

Obrution Bed

The Coenothyris Bed in the lower part of subunit 31 of the Saharonim Formation (figs. 3, 4) can best be characterized as an obrution bed; that is, very rapid burial of intact brachiopods that resulted in exceptional preservation. A rapid burial event of this type can record brief moments in time or “snapshots” of seafloor communities (sensu Brett, 1995; Taylor and Brett, 1996; Brett et al., 1997) and may form important marker horizons in event stratigraphy (Brett et al., 1997). Ager (1974) noted that the characteristics of tempestites, or storm deposits, are that they show evidence of violent disturbance of preexisting sediments followed by their rapid redeposition, all in a shallow water environment. Here, however, there is no indication that preexisting sediments were disturbed. In any case, Ager's (1974) definition for tempestites may be too restrictive in demanding evidence for violent disturbance of preexisting sediments. Mud tempestites may be a more appropriate term for indicating the rapid settlement of sediment in relatively low energy settings after a storm (Brett, personal commun.). The Coenothyris obrution bed would seem to fall into this category. In most cases there is no direct evidence the preexisting sediments were disturbed, only evidence of rapid redeposition. It is probable that in most cases redeposition was involved. The only other case might be flood washout of muds following the flooding of rivers, but these would be a form of storm deposit in most instances. The brachiopods here need not have been disturbed if they lie below storm-wave base; they will still get the backwash of storm-suspended sediments. The evidence here does suggest deposition below storm-wave base. Stanley (1994) illustrated terebratulid brachiopods in the Early Mesozoic Shalypayco Formation of northern and central Peru that resemble the Coenothyris Bed in richness but differ in that the orientations and packing suggest storm deposition.

The burial of immature Coenothyris in the Saharonim Formation best conforms to Johnson's (1960) model 1 in which a life assemblage is buried suddenly with no evidence of postmortem transport or exposure (see also Johnson, 1997: fig. 7.7). This community can best be categorized as having a taphonomic grade of “A” based on Brandt's (1989) scheme in which there is <10% fossil breakage, >90% articulation, and <10% corrasion/orientation. Johnson's (1960) model 2 describes a more gradual accumulation and burial of shells winnowed by brief postmortem exposure and transport (see, for example, Stanley's [1994: fig. 17] illustration of a bed of Upper Triassic terebratulids where orientations and packing suggest storm deposition; note the high proportion of disarticulated valves), whereas model 3 describes an assemblage altered by long postmortem exposure and appreciable transportation from the original habitat.

The Coenothyris Bed does not seem to be hiatal (condensed) or a lag accumulation, as shells are all well preserved and it is not traceable laterally for any great distance. Brett and Baird (1993) considered that a first line of evidence for hiatal concentrations is stratigraphic and that in passive continental margins and epeiric seas these beds display a remarkable lateral persistence, even across major changes in facies of the adjacent beds, although so too are some obrution beds. However, not only condensed beds may show large lateral persistence. Brett and Baird (personal commun.) have traced out many single event obrution beds in the Ordovician–Silurian and Devonian with considerable precision. Indeed, the fact that this occurrence appears localized may require explanation. They suggested (unpublished manuscript) that two mechanisms of mud redeposition may be indicated: (1) widespread layers may indicate hemipelagic sedimentation from detached sediment plumes, and (2) localized occurrences may reflect point source bottom flows (true mud turbidites or distal tempestites).

There is also no evidence of mixtures of fossils representing two or more usually distinct biofacies as described by Kidwell and Bosence (1991) on the obrution horizon (although there is shell mixing in the subsurface horizon), nor is there any indication of corroded or abraded shell fragments typical of environmentally condensed shells beds and simple composite beds. All of the brachiopods are articulated, although not all are in life position. Further indication that the beds are not lag deposits is the lack of any reworked fossil material derived by significant erosional truncation of older strata (Kidwell, 1991).

Seilacher (1990) noted that some groups are more susceptible to catastrophic smothering than others by virtue of their general organization and that sudden mud sedimentation can kill and preserve victims simultaneously. He also mentioned that the echinoderms are one such group because their ambulacral system communicates with the ambient sea water and becomes easily clogged by fine sediment. Although some brachiopods have good sediment shedding mechanisms, they may constitute another such group since they are extremely sensitive to suspended sediment clogging the lophophore. Also, they are sessile and therefore unable to escape burial as can trilobites and crinoids. Datillo (2004) reported on trace fossil evidence that the plectambonitoid brachiopod Sowerbyella rugosa from the Ordovician of Kentucky was able to move upward through the sediment to reach the sediment–water interface by snapping its valves. He further noted that if plectambonitoids were able to snap their valves to move through the sediments, strophomenoids might have been able to accomplish the same maneuver. Here, however, the biconvex Coenothyris shells differ morphologically from the concavo–convex-geniculated Sowerbyella rugosa and were unlikely to have been able to either shed or move through the sediment. If the Coenothyris brachiopods were nonpedunculate and able to have assumed a near-vertical, semiinfaunal position to escape catastrophic burial as were the Ordovician Sowerbyella shells, they might have escaped burial by “valve snapping”.

Coenothyris shells preserved are in perfect condition, with >99.9% of the shells articulated and only three free valves evident (out of 2399 shells studied in the obrution bed). The shells lay directly on the buried surface, in this case a bivalve pavement, and were buried suddenly. Johnson (1989) described an exterminated Pentamerus population from the Silurian of Norway that was terminated not solely by the physical disruption of the sea by storm waves, but also by passive suffocation by fine clastic sediments falling out of suspension. Johnson's populations tended to have a high percentage of whole individuals, as is the case with the Coenothyris Bed described herein. He concluded that the termination of an immature population was due to an exceptionally severe storm.

Obrution Event in Relation to Depositional Sequence

The obrution deposit represented by the accumulation and preservation of Coenothyris occurs in the lower part of the Fossiliferous Limestone Member, Saharonim Formation. Brett and Seilacher (1991) noted that obrution deposits are most frequently recognized in the late transgressive to early highstand deposits of many successions. According to Druckman (1974b), the Fossiliferous Limestone Member was deposited as part of the transgression that prevailed in the area during Upper Anisian to Lower-Middle Ladinian times. Brett and Seilacher (1991) noted that this empirical relationship is somewhat paradoxical because transgressive sections are relatively the most sediment starved and consequently should contain the least number of burial events. They noted that many condensed sections, however, may accumulate as a stack of very infrequent large depositional events and that major storms may carry siliciclastic or allodapic carbonate sediments into basins that are otherwise almost completely sediment starved. Reworking is not a factor in the preservation of organisms trapped in these deposits because of the deep position of the sea floor. The Coenothyris Bed under discussion here is an example of a sediment-starved shell bed that was smothered.

Allochthonous Versus Autochthonous Burial

Based on examination of the brachiopod and bivalve fauna preserved in the Coenothyris Bed, there is no evidence of preferential alignment of valves or hydrodynamic sorting (see figs. 3–5), two criteria noted by Brett and Baird (1993) for recognizing, at least locally, some degree of current reorientation. The elongated shape of the epifaunal Modiolus would make that bivalve shell a sensitive indicator of any current that would most likely result in at least a weak alignment of shells. The infaunal bivalves (Pleuromya, Myophoria) are preserved articulated, but with few specimens in burrows. However, in two cases, both observed in the field, these infaunal bivalves were buried within their burrows, thus ruling out postmortem transport (Brett and Baird, 1993; Kranz, 1974). Most Coenothyris shells (89.9%) are preserved in life orientations. Moreover, Alexander (1986) provided evidence that the decay of brachiopod pedicles after death would result in postmortem instability unless they were buried very rapidly. Thus, the obrution bed represents an autochthonous deposit, that is, a deposit of taxa buried in their habitat.

Environment of Deposition

Druckman (1974b) noted that the occurrence of a diverse benthic and nektic macrofauna, along with the absence of evaporites and dolomites, suggests that deposition of the Saharonim Formation occurred in an open-shelf environment of normal salinity. He further suggested that the complete absence of scouring within the carbonates or signs of channeling and ripple marks implies that most of the member was deposited at least beneath wave base (at a paleolatitude of within 10°N) and may have been deposited even at a depth of between 100 and 200 m. Westermann (1996) suggested that the cephalopod Monophyllites (common in the lower part of the Saharonim Formation although not recovered from the Coenothyris Bed) indicates a deep ocean environment. However, the marine Middle Triassic series of Israel belongs to the Sephardic Province that sharply differs in its cephalopod composition from coeval “normal” or Panthalassan faunas. Extreme abundance and diversity of nautiloids as in the Triassic of Israel are diagnostic of shallow marine environments according to Bucher (personal commun.). Bucher further noted that the Sephardic Province is also known from Spain and shows some affinities in its abnormal faunal composition and shallow depositional environments with the Germanic Muschelkalk (fig. 6A–C). Triassic phylloceratids such as Monophyllites are not, according to Bucher, indicative of deep water, but are generally ubiquitous in ammonoid-bearing beds. In Israel, cephalopods are common only in the deeper more transgressive basins where there is an opening to the Tethys.

Simple layered bottom horizons may be sequentially arranged along a proximal-distal axis, and simple layers, of which the Coenothyris Bed is typical, are more common in proximal upslope settings (Brett et al., 1986). However, concretionary and pyritic beds are typical of more distal downslope environments. The depth of the Coenothyris Bed may to be similar to that inferred for Johnson's (1989) Pentamerus beds from Norway. He noted that suffocated populations tend to have a high percentage of whole individuals and that whether by fallout below or active disruption above wave base, only the most severe storms appear to have affected these deeper, more offshore sea-floor zones.

The burial increment (sensu Kidwell, 1991) consists of unfossiliferous sediment ranging in thickness from 1.5 to 2 cm that was rapidly deposited in a period that ranged from a few hours to a few days. There is no evidence of fining-upward size gradation or planar and cross lamination that are indicative of deposition from a waning current (Brett, 1990). Brett noted that the burial layer most commonly consists of barren structureless mudstone or siltstone. Here it consists of marl in the lower part and pure clay in the upper part with no evidence of gradation between the two layers. The most likely reason for rapid burial here was that the muds were probably flocculated, causing the suspended particles to behave hydrodynamically as silts or fine sands rather than as muds, as discussed by Brett (1990) and Brett and Taylor (1997). This seems most plausible given the high clay content of the sediment (22% clay in the lower part of the bed, with the remainder consisting of carbonate, no pyrite in the clay fraction, no quartz, mica, or silt, and no indication of volcanics; 100% clay in the upper part, no carbonate, no fine sand or silt). In fact, clay matrix can be seen filling the spaces between the immature brachiopod shells (fig. 7A) in the upper part of the bed, but it is not clearly visible on the sectioned slabs (fig. 7B, C) due to the crumbly nature of the claystone; it spalled off during sectioning. As noted by Brett (1990), most of the best obrution Lagerstätten are preserved in and beneath layers of fine-grained sediments, for which extremely rapid sedimentation might seem an impossibility.

A small percentage (10.1%) of the shells (of 2336 counted on bedding surfaces) are inverted, dorsal side up (see appendix), signifying some kind of physical perturbation of the environment during or prior to burial. Taylor and Brett (1996) indicated that environmental disturbances associated with the onset of a storm may have produced changes in salinity, oxygenation, and turbidity that adversely affected the benthos. However, most Coenothyris are in a horizontal to subhorizontal attitude relative to the sediment– water interface, ventral side up, indicating clogging of the mantle and subsequent collapse and decay of the pedicle before final burial. Webby and Percival (1983) reported rapid decay after displacement of the soft parts of Eodinobolus, an inarticulate with no functional pedicle present during ontogeny (Norford and Steele, 1969), from the Ordovician of New South Wales. Here, however, decay of the pedicle occurred before displacement; otherwise, the pedicles would have kept the brachiopod shells attached in a vertical (= living) position. As noted by Boardman et al. (1987), the presence of a pedicle foramen in fossil brachiopods does not necessarily mean that the shell was attached to the substrate by the pedicle. They further stated that there are some individuals of living species that are attached by their pedicles when hard substrates are available, but are able to adapt a free-lying mode of life when only mud bottoms are present. Here, however, attachment was on a hard bivalve substrate.

In this case Coenothyris was smothered by a sudden pulse, or pulses, of clay sedimentation, as a sedimentary event deposit, and subsequently subjected to weak currents. This explains the small percentage of articulated shells dorsal side up. Druckman (1974b) proposed that cyclic alternation of fossiliferous limestone and shale layers on a meter scale in the Lower Saharonim Formation may be due to pulses of terrigenous clay that were supplied to the area over a steady “background” of carbonate precipitation or secretion. However, it seems possible that: (1) there was a change in salinity, and/or (2) there was a change in oxygen levels, and (3) there was an episode of sedimentation that clogged the mantles of the brachiopods resulting in death by anoxia. This was followed by a period of weak current activity that flipped about 10% of the dead shells whose pedicle foramens were either weakened or decayed.

Some shells were buried rapidly, as evidenced by the layer of spar-filled brachiopods (see the lower half of fig. 7B) since rapid entombment does not allow for the decay of soft tissues with subsequent displacement by sediment. Also, the decay of organic matter may increase alkalinity and favor early mineralization. The brachiopods just above the spar-filled zone (fig. 7B, upper half) are mud filled, indicative of individuals that had died earlier and decayed to become infilled with sediment; approximately 25% of the shells were spar filled. There are also some shells with geopetal structure (fig. 7B) that indicates the original orientation of the rock, with the spar-filled upper half representing the “up” position and the spar–mud boundary roughly parallel to the sediment– water interface.

Bioturbation by the burrowing bivalves, mainly by the relatively large Pleuromya and the medium-sized Myophoria, likely overturned the brachiopods after death. Well-preserved bivalves that formed the burial pavement were probably smothered by a previous influx of sediment too thick to allow them to escape. However, some of the infaunal forms (e.g., Pleuromya, Myophoria) may have been able to escape death and bioturbate the sediment after the settling of the Coenothyris spat. There is no other evidence of disturbance among Coenothyris, even minimal disturbance, similar to what Taylor and Brett (1996) reported for their low-level benthic forms (e.g., brachiopods) that were toppled and inverted with minor disarticulation prior to burial. There are no signs of current activity, no ripple marks, no scours of any kind, and no significant numbers of disarticulated shells (3 out of 3000); there is no indication of current activity strong enough to turn the shells upside down. Less than 1% of the bivalves that occur with Coenothyris are open, facing down into the sediment—an indication of death position. Myophoria valves can be observed (fig. 7B, C) in a concave-down position as a disarticulated hash, but this represents the pavement upon which the Coenothyris spat settled and lies directly under the obrution bed. The bed as a whole is time-averaged (see fig. 7C) and includes pavements of bivalve shells that were reworked and stacked. This process facilitated colonization by pedunculate brachiopods in providing a hard substrate for attachment. The shell bed may represent an interval of sediment starvation.

Community and Population Structure

The Coenothyris population is composed of juvenile brachiopods that were rapidly buried after they settled on a shell pavement that consisted of 10 bivalve genera. Whereas there is clearly evidence of time-averaging, which is an assemblage in which the individuals preserved together were not alive at the same time (Flessa, 2003), the community under study here is represented by both the brachiopods and the bivalves (which were likely buried earlier than the brachiopods) found in the lowermost beds of subunit 31. The Coenothyris Bed is considered a local paleocommunity since the shelly parts of the original community were collected from a single bed at one outcrop that belong to within-habitat, time-averaged assemblages (Zuschin and Stanton, 2002). Some of the infaunal bivalves likely escaped immediate death by burrowing. The brachiopods represent a single cohort (table 2) of juvenile mortality of one spatfall. Some workers (e.g., Cloud, 1948) recognized that environmental factors could have produced a subnormal life-span among brachiopod populations. There is no evidence of stunting; the environment was well oxygenated, salinity was normal, and nutrients were abundant, as evidenced by a diverse fauna above and below the obrution bed (see table 3). The shells were preserved in a type 3 deposit (Brett et al., 1997), which is the most common type of obrution deposit, mud tempestites, which are typically preserved in lower-energy settings.