#bromacker quarry

Holotype specimen of Tambacarnifex unguifalcatus, preserved in couterparts. Photographs by Dave Berman, 2010.

Tambacarnifex unguifalcatus was discovered by Thomas Martens and his father Max in 1995 in the same pocket of fossils from which the first-discovered specimen of the herbivorous basal synapsid Martensius bromackerensis was recovered. Because numerous fossil animals were jumbled together, Thomas and Max weren’t able to collect individual specimens from the bone pocket using our standard technique of surrounding a specimen in a plaster and burlap jacket. Instead, they collected all the individual pieces of rock that contained bone or at least appeared to contain bone, as most rock pieces were coated in goopy mud. Thomas cleaned the rock pieces with water to reveal the bone, and then pieced together the various specimens.

He eventually sent us the specimen that became the holotype of Tambacarnifex, along with pieces that he thought might go with it. Dave and I spent hours piecing together the remainder of the skeleton, and we searched the collections at the Museum der Natur, Gotha for missing pieces in subsequent field seasons. The majority of the specimen was recovered, but the skull, a few vertebrae, and distal finger and toe bones are missing. A rock piece with the greater portion of a lower jaw with teeth was also collected from the bone pocket, though it couldn’t be associated with the skeleton and may represent a second individual. A lot of bone was lost from the specimen, but impressions of missing bone were preserved, which proved useful for identifying wrist and ankle bones, among others. Dave used a white pencil to color in the bone impressions so they would stand out for study and in photographs of the specimen. Ultimately, we realized that Martensius and Tambacarnifex were preserved one on top of the other, though separated by several inches of rock.

The lower jaw piece of Tambacarnifex unguifalcatus. Photograph by Dave Berman, 2008.

The teeth of Tambacarnifex preserved in the lower jaw are strongly recurved and flattened side-to-side, which, along with other features preserved in the skeleton, indicate it is a member of the basal synapsid group (family) Varanopidae and in the subfamily Varanopinae. The Varanopidae have been likened to the actively predaceous modern monitor lizards in the family Varanidae, hence the similar name. Varanopids were the most diverse and longest-surviving basal synapsids, being known from the Late Carboniferous–Middle Permian (~309–260 million years ago) of North America, Europe, Asia, and Africa. With their sharp, recurved teeth and a gracile skeleton, scientists think varanopids were agile predators, at least compared to other animals of their time. They range from about 12–78 inches in length, with the smallest ones probably being insectivorous and the larger ones carnivorous. Tambacarnifex has an estimated body size of about 35 inches, and as a medium-sized varanopid with gracile limbs it would have been an agile carnivore, preying on on any of the Bromacker vertebrates that it could catch.

An articulated but incompletely preserved series of 11 vertebrae of Tambacarnifex unguifalcatus. Notice that the neural spines are low and subrectangular, so it is unlikely that they supported a sail, as occurs in some other basal synapsids such as Dimetrodon teutonis. The front of the animal is to the left. Photo by Dave Berman, 2008.

Unlike Dimetrodon teutonis, the other apex predator at the Bromacker, Tambacarnifex has broad, low neural spines that alternate in height. It differs from other varanopines in the shape and anterior inclination of its neural spines and in having greatly elongated and recurved bony claw supports in its hands and feet. The generic name Tambacarnifex was coined in reference to its position in the food chain: “Tamba,” for the Tambach Basin, which the holotype inhabited, and the Latin “carnifex,” meaning executioner, for its role as an apex predator. “Unguifalcatus” was derived from the Latin “unguis,” nail or claw, and “falcatus,” meaning sickle-shaped, in reference to the long, strongly recurved bony claw supports.

Incomplete front (left) and hind (right) feet of Tambacarnifex unguifalcatus. Notice the extremely long bony claw supports preserved on the first, third, and fourth fingers of the front foot and the fourth toe of the hind foot. I–V refer to finger and toe numbers. Photos by Dave Berman, 2008.

Illustration of Tambacarnifex unguifalcatus consuming a Dimetrodon teutonis carcass. Outline drawing by Matt Celeskey, colored (with permission) by Carnegie Museum of Natural History Vertebrate Paleontology Scientific Illustrator Andrew McAfee.

Stay tuned for the final post of this series, which will summarize what we’ve learned about the Bromacker. Click here if you would like to download your own copy of the outline drawing of Tambacarnifex consuming Dimetrodon to color in. The paper describing Tambacarnifex unguifalcatus can be viewed by clicking here.

Holotype specimen of Dimetrodon teutonis, which consists of a partial vertebral column. The preserved portion of this vertebral column is highlighted in the reconstruction of Dimetrodon (lower right). Photograph by the author, 2007. Dimetrodon reconstruction modified from Romer and Price, 1940.

Specimens of two top predators have been discovered at the Bromacker quarry. Like Martensius, both are basal members of the group Synapsida, the later members of which gave rise to mammals. You might be familiar with one of them – Dimetrodon, a synapsid sometimes incorrectly portrayed with dinosaurs, which carried a tall sail on its back that was supported by bony spines. The other is a new genus and species that will be presented in my next post.

The fossil pictured above, the first-discovered specimen of Dimetrodon from the Bromacker quarry, may not look like much, but it was the first record of Dimetrodon outside of North America. The circumstances under which it was found were very different from the discovery of other fossils from the Bromacker quarry. Before Dave Berman and I arrived for the 1999 field season, Thomas Martens noticed that someone, possibly a fossil poacher, had been in the quarry overnight and knocked some rocks off the quarry lip. The rocks apparently broke upon hitting the ground, which exposed some bones. Thomas carefully picked them up and took them to his lab at the Museum der Natur, Gotha (MNG). When Dave and I met Thomas at the quarry on our first day of the field season, Thomas mentioned the find and told us that he thought the bones were ribs. We didn’t think much of it, other than horror at learning a fossil poacher might have visited the quarry overnight, one of our worst fears.

As planned, Dave and I spent the last day of the field season in the museum collections, and when Thomas let us in that morning, he reminded us to look at the potential ribs and told us where they were. Shortly after we began examining them, Dave and I simultaneously realized that the “ribs” were actually spines of Dimetrodon. We couldn’t believe our eyes, because of all the Early Permian fossils known from North America, Dimetrodon was Thomas’ favorite. Indeed, he’d used an image of it on signs at the Bromacker and included a model of Dimetrodon in a diorama, once on display in the MNG, that showed models of Bromacker animals in their environment. Thomas jumped for joy later that day when we gave him the news.

So how did Dave and I so quickly realize that the “ribs” were spines of Dimetrodon? Besides Dimetrodon, some other basal synapsids had sails, the function of which remains unknown, though scientists have speculated they could’ve been used for display or regulating body temperature. The spines (known as neural spines) supporting the sails vary in shape and length, with those of Dimetrodon and its herbivorous relative Edaphosaurus being tall and narrow, and those of another relative, the carnivorous Sphenacodon, being shorter and blade-like. Neural spines of Dimetrodon are easy to distinguish, because in addition to being long they bear fore and aft grooves, which create a dumbbell-shaped cross-sectional outline, and they lack the ‘crossbars’ that occur on the long neural spines of Edaphosaurus. When Dave and I saw the fore and aft grooves, the dumbbell-shaped cross-sectional outline of some broken spine ends, and an absence of crossbars, we knew that the “ribs” were indeed spines of Dimetrodon.

Flesh reconstructions of Sphenacodon sp. (left) Dimetrodon grandis (middle) and Edaphosaurus pogonias (right) to show the differences between their sails. Note that Dimetrodon and Sphenacodon are more closely related to one another than they are to Edaphosaurus, despite their different sail shapes. Reconstructions of Sphenacodon and Dimetrodon by Dmitry Bogdanov and that of Edaphosaurus by Nobu Tamura, all from Wikimedia Commons.

The Bromacker Dimetrodon is considerably smaller than other known species of the genus, and this is one character among other more detailed anatomical features that distinguishes it. For the new species name, Dave selected the Latin “teutonis,” which means an individual of a German tribe, in reference to the geographic origin of the holotype specimen.

Two additional specimens of Dimetrodon teutonis. Left, hindleg and shoulder girdle bone (fused scapulocoracoid) and right, several vertebrae bearing complete to nearly complete neural spines of an individual that was larger and presumably more mature than the holotype. Photographs by the author, 2007.

Dave was able to use a mathematical equation involving measurements of the vertebrae to estimate the holotype’s weight as a living animal at 31 pounds. In contrast, other known Dimetrodon species have estimated weights of about 81–550 pounds. We later discovered additional partial specimens of Dimetrodon at the Bromacker quarry, and Dave estimated the weight of the largest specimen with vertebrae at 53 pounds, still considerably less than that of what had previously been the smallest species, D. natalis from Texas. Dimetrodon is otherwise known from numerous species from the American mid-continent and southwest that generally got larger through time.

Reconstructions of various species of Dimetrodon drawn to scale. The diminutive D. teutonis is at bottom center and D. natalis, no longer the smallest species, is at bottom left. Illustration adapted from Dmitry Bogdanov via Wikimedia Commons. 

All Dimetrodon species have teeth adapted for meat-eating in being teardrop-shaped with sharp edges for slashing flesh. By size and jaw position these sharp teeth are divided into precanines, canines, and postcanines of varying numbers. Unlike D. teutonis, some species even had fine serrations on their tooth edges. The only known upper jaw bone of Dimetrodon teutonis clearly has two canines, but one is missing and represented by a large gap in the tooth row that would have accommodated this tooth. The second canine is represented only by its broad base, but it too must have been large. Although it was a small animal, the teeth of D. teutonis indicate that it was a meat-eater and as such would have preyed on other vertebrates from the Bromacker, many of which were even smaller.

Diagrammatic drawing of the skull of Dimetrodon (left) and photograph of the maxilla or upper jaw bone (right) of D.teutonis. Abbreviations: c, canine; pc, postcanine; prc, precanine. Photographs by the author, 2007. Drawing of skull from Wikimedia Commons.

Stay tuned for my next post, which will be about the second-known apex carnivore from the Bromacker. In the meantime, here are links to scientific papers on Dimetrodon teutonis:

https://www.researchgate.net/publication/325670232_A_new_species_of_Dimetrodon_Synapsida_Sphenacodontidae_from_the_Lower_Permian_of_Germany_records_first_occurrence_of_genus_outside_of_North_America

https://www.researchgate.net/publication/288544821_New_materials_of_Dimetrodon_teutonis_Synapsida_Sphenacodontidae_from_the_Lower_Permian_of_Germany

Adult, holotype specimen of Martensius bromackerensis. Image digitally assembled by the author from five photographs taken by Diane Scott (Preparator at University of Toronto Mississauga [UTM]), 2010–2013. The specimen was collected in several large blocks.

The formal publication of some of the Bromacker discoveries took more time to complete than others, and our most recently pubished fossil, Martensius bromackerensis, holds the record in that regard. Four nearly complete specimens of Martensius were collected from the Bromacker quarry between 1995–2006. The first, discovered by Thomas Martens and his father Max, came from a jumbled pocket of fossils. Unfortunately, muddy groundwater had penetrated cracks in the subsurface of this portion of the quarry and coated and eroded bone present along these cracks. Despite this damage and the lack of a skull, we could identify the specimen as a caseid synapsid (synapsids, also known as mammal-like reptiles, are a group of amniotes whose later-occurring members gave rise to mammals).

Drawing of 1995 Martensius bromackerensis specimen. Because the specimen was collected in numerous pieces of rock, with parts of some bones exposed on apposing rocks, Scientific Illustrator Kevin Dupuis (UTM) had to first draw the bones exposed on each piece and then assemble all of the drawings digitally. Dotted lines indicate bone impression in the rock. Arrows point to healing scars from two fractures in the last right rib. Additional healing scars can be seen in preceding ribs. This animal apparently survived a serious injury. Modified from Berman et al., 2020.

The next specimen was discovered in 1999 by Georg Sommers (Preparator, Museum der Natur, Gotha), who prepared the fossil. It consists of a vertebral column, ribs, some limb bones, and a few scattered skull elements. Unfortunately, a more complete skull was needed to allow for comparison to other caseids, some of which are based only on skull material. It wasn’t until the discovery of two more specimens in 2004 and 2006 by Stuart Sumida and Dave Berman, respectively, that the long sought-after skull was found. Preparation of these specimens took a long time due to their size and the considerable amount of rock covering the bones in some of the blocks. My promotion to Collection Manager in 2005 left me with considerably less time to prepare fossils. Other preparators were asked to help with the preparation at both Carnegie Museum of Natural History (CMNH, Dan Pickering and Tyler Schlotterbeck) and in Dr. Robert Reisz’s lab at the University of Toronto at Mississauga (Diane Scott and Nicola Wong Ken). Robert was originally slated to lead the study, but other commitments prevented him from working on it, so Dave took over.

Besides preparation, the scientific study and publication of the specimens required illustrations and photographs, most of which were done by Diane, Nicola, and Kevin. Andrew McAfee (Scientific Illustrator, CMNH) made skeletal and flesh reconstructions of the animal, as well as an illustration of two Martensius in their ancient habitat (see The Bromacker Fossil Project Part III for a link to this illustration). All of this effort was worth it, however, because besides adding to the diversity of the Bromacker vertebrate fauna, Martensius has an unusual life history.

Juvenile specimen of Martensius bromackerensis. Image digitally assembled by the author from two photographs (skull and body) taken by Diane Scott in 2013. The skull, shown in ventral aspect, is incomplete and eroded on its dorsal surface.

Caseid synapsids are a diverse, long-lived group known from the Late Pennsylvanian–Middle Permian epochs (~300–259 million years ago) of Europe, Russia, and the USA, and, with one exception, all are adapted to eating plants (herbivorous). The most advanced caseids (such as the enormous Cotylorhynchus romeri) have ridiculously small skulls when compared to those of carnivores, spatulate (spoon-shaped) teeth tipped with small tubercles (cuspules) for cropping vegetation, and huge, barrel-shaped ribcages to support a large gut for fermenting cellulose-rich plants. The exception is the earliest known (Late Pennsylvanian epoch, ~300 million years ago) caseid, Eocasea martini, represented by a single, incomplete juvenile specimen from Kansas. The teeth of Eocasea are small and conical, which indicate that it most likely ate insects. Because it’s skull and ribcage are of normal size, in contrast to juveniles of Martenius, Eocasea probably ate insects throughout its life.

Reconstruction of the skull of Martensius bromackerensis (left) from the Early Permian (~290 million years ago) Bromacker quarry, Germany, and the more advanced caseid Ennatosaurus tecton (skull, middle and skull fragment with cuspule-tipped teeth, right), from the Middle Permian (~263 million years ago) of Russia. Skull reconstruction of Martensius made by Diane Scott and modified from Berman et al., 2020. Ennatosaurus skull reconstruction and jaw fragment drawing modified from Maddin et al., 2008.

Martensius has a modestly expanded ribcage and a small skull, suggesting that it was herbivorous. Furthermore, the feet of Martensius, like those of other caseids in which the feet are known, are large, with massive, elongated, strongly recurved claws. Martensius also has a well-supported hip region that may have enabled it to rise on its hind legs to reach and tear down overhead branches to feed upon.

The upper and lower teeth of the adult Martensius differ from those of more advanced caseids in being triangular and lacking cuspules. The upper jaw teeth of the juvenile resemble those of the adult, but the lower jaw teeth are more numerous—31 in the juvenile compared to 25 in the adult—and surprisingly, they resemble those of Eocasea. Dave concluded that juveniles of Martensius had teeth adapted for eating insects, which were replaced by an adult dentition that would’ve been good for cropping plants and piercing insects. Remarkably, the juvenile Martensius apparently died while in the process of replacing its juvenile dentition with that of adults.

So why have different juvenile and adult dentitions? Modern animals that eat fibrous plant matter have micro-organisms called fermentative endosymbionts in their large guts, which break down difficult-to-digest plant matter via fermentation. It is assumed that early fossil plant-eaters with broad ribcages also had large guts housing fermentative endosymbionts. Prior to the discovery of Martensius, other scientists hypothesized that early herbivores acquired endosymbionts by eating herbivorous insects that already had these microbes in their guts. In Martensius, the introduction of endosymbionts apparently occurred during the juvenile, insectivorous stage of life, which set the stage for adults to add plants to their diet.

Flesh (top) and skeletal (bottom) reconstructions of Martensius bromackerensis. Illustrations by Andrew McAfee and modified from Berman et al., 2020.

The generic name Martensius honors Thomas Martens for his discovery of vertebrate fossils at the Bromacker quarry and his perseverance in maintaining a highly successful, long-term field operation resulting in the discovery and publication of the exceptionally preserved Bromacker fossils. Bromackerensis refers to the Bromacker quarry, the only locality from which this species is known.

Stay tuned for my next post, which will feature some terrestrial dissorophoid amphibians.

For those of you who would like to learn more about Martensius, here’s a link to the 2020 Annals of Carnegie Museum publication in which it was described: https://www.researchgate.net/publication/340309382_New_Primitive_Caseid_Synapsida_Caseasauria_from_the_Early_Permian_of_Germany

Collage of the fossils highlighted in this series. Images not to scale. Photos by the author, Dave Berman, and Thomas Martens.

The Bromacker quarry is a rare site in that it preserves exquisite, articulated fossils of a unique vertebrate fauna that lived in an atypical or rarely recorded Early Permian (~290 million years ago) setting. Early in our work at the Bromacker, we became aware that the fossil vertebrates we were finding were unknown or extremely rare in Europe but were closely related or identical to species commonly found in North America. Until then, most of the fossil vertebrates found in Europe were discovered in gray to black sediments deposited in ancient lake beds, whereas the fossils from the Bromacker quarry occurred in red beds representing a terrestrial setting. Paleontologists looking for fossils in Europe typically prospected the gray to black sediments where fossils were relatively plentiful rather than red beds, which were thought to represent arid environments not conducive to fossil preservation.

Photograph of a diorama showing the Tambach Basin 290 million years ago, which was once exhibited at the Museum der Natur, Gotha. It was built in 1996, so many of the inhabitants of the basin weren’t yet discovered. One of these is Dimetrodon teutonis, which was inadvertently depicted as being large and numerous. Image provided by Thomas Martens, 2020.

In a collaborative effort to help determine how the fossil deposit at the Bromacker quarry formed and why its vertebrate fauna is unique, Dave Berman invited his colleague David Eberth to join us for the 1998 field season. David is a geologist/vertebrate paleontologist who was then employed by the Royal Tyrrell Museum of Palaeontology in Canada but is now retired. The sediments preserving the Bromacker fossils are part of a rock unit called the Tambach Formation and were deposited in the Tambach Basin. The results of David’s study, built in part upon investigations by other geologists and paleontologists, indicate that the Tambach Basin was situated within an ancient mountain range and isolated from river systems. At the time the fossils were deposited, the basin was internally drained, and as a result, when it rained, water would flow towards the basin center and form ephemeral ponds and lakes. Based on the geology, fossil plant assemblage, and geographic setting of the Tambach Basin, David concluded that the climate was possibly similar to the wet‑and‑dry tropical climate of modern North African savannas, Brazilian Campos, or the Venezuelan Llanos.

Map showing the areal extent of the Tambach Formation today and the inferred boundary of the Tambach Basin, with arrows indicating direction of water flow. The northern boundary of the basin is not preserved, but it was thought to have been closed when the Bromacker fossil deposit formed. Modified from Eberth et al., 1997.

Most of the fossils discovered at the Bromacker quarry came from two massive units, the more fossiliferous of which is about 21 inches thick, that formed in separate major flooding events. David theorized that these deposits formed when heavy rain caused a sheet-flood of sediment‑laden water to sweep down the sides of the Tambach Basin and across the basin floor, killing any animals that couldn’t escape the flow. The sheet-flood transported the carcasses to the basin center where they were deposited, rapidly buried, and eventually fossilized. These deposits record a unique snapshot of vertebrate life in the Tambach Basin, because only animals inhabiting the basin would have been captured by the sheet-flood.

In contrast, most Early Permian fossil‑bearing deposits in North America formed on coastal or alluvial plains. Carcasses would’ve been transported to the deposition sites by rivers, some of which had a large geographic reach. These types of deposits can accumulate over a long period of time and have potential to mix together fossils from different environments.

Photograph of a diorama once exhibited at Carnegie Museum of Natural History that shows a typical Early Permian peat swamp or backwater swamp of a major river system. A similar modern environment would be the Okefenokee Swamp, Georgia. Photo by Mindy McNaugher, 2007.

Besides having an atypical geographic setting, the makeup of the Bromacker vertebrate fauna differs from those known from other Early Permian sites. The Bromacker vertebrate fauna has a low diversity of terrestrial tetrapods, but more importantly, it lacks fishes and aquatic to semi‑aquatic constituents. This is probably due to the Tambach Basin’s isolation from regional river systems and because it experienced seasonal to sub‑seasonal drying, making it difficult for water-reliant vertebrates to become established. Based on numeric counts of individual specimens, we determined that the relatively large‑sized herbivores Diadectes, Orobates, and Martensius greatly outnumbered the synapsid apex predators Dimetrodon and Tambacarnifex. We think the rarity and low diversity of synapsid carnivores is probably due to the lack of an aquatic to semi‑aquatic component in the food chain.

In contrast, most Early Permian North American localities preserve a diverse, mixed aquatic‑terrestrial fauna that either lived in water or was closely associated with water and aquatic food chains. Herbivores were rare in terms of both diversity and numbers, whereas synapsid apex predators were diverse and numerous.

A more dynamic North American Early Permian scene that includes a mixed aquatic‑terrestrial vertebrate fauna. The Dimetrodon on the right has caught a freshwater shark, demonstrating the importance of aquatic animals in the food chain. © Julius Csotonyi/Houston Museum of Natural Science.

The Bromacker is the oldest known terrestrial vertebrate ecosystem in which herbivores greatly outnumber apex carnivores, and in that respect, it resembles terrestrial vertebrate ecosystems of today. A modern example is the African savanna in which large herds of herbivores such as zebra, wildebeest, and buffalo provide a food source for a much smaller number of carnivores including lions, cheetahs, and hyaenas. Indeed, we consider the Bromacker to represent an early stage in the development of the modern terrestrial vertebrate ecosystem and that these early stages were restricted to upland areas isolated from aquatic‑based food chains.

This summary concludes the Bromacker Fossil Project blog post series. I hope that you’ve enjoyed reading it. Cast replicas of many of the fossils described in this series are exhibited in the Fossil Frontiers display case in CMNH’s Dinosaurs in Their Time exhibition, so be sure to look for them on your next visit. I’m grateful to Dave Berman, Albert Kollar, Thomas Martens, and Stuart Sumida, who answered numerous questions and provided photographs, and to Patrick McShea and Matt Lamanna for their editing skills. Click here to read the paper by Eberth et al. 2000.

This post will be the first of a series focusing on notable fossil animals discovered in the Bromacker quarry. I selected Diadectes abisitus, a member of the ancient group Diadectomorpha, to present first because, had it not been discovered in the first year of the collaborative field work, the project might not have continued.

Dave Berman and his colleague Stuart Sumida (California State University, San Bernardino) joined Thomas Martens for five weeks of field work in the summer of 1993. They dug a quarry over six feet deep in their search for fossils, while working in a mix of hot and humid or near freezing temperatures, with plenty of rain. It wasn’t until the second-to-last day of the field season that the Diadectes specimen was discovered. By then, as Dave later told me, he was so discouraged by the lack of fossils that he assumed this would be his first and last field season in Germany. I should mention that Stuart had previously uncovered a few small vertebrae, but because the vertebrae resembled an animal described from the Bromacker in 1991, the team was not very excited about the discovery. They couldn’t have been more wrong in their field identification of the vertebrae, however, but more on that in a later post.

The 1993 quarry shortly before discovery of Diadectes absitus. Pictured are Stuart Sumida standing in the quarry and Thomas Martens crouched to his right. The Diadectes fossil was found in the corner opposite Stuart’s left shoulder, which is out of sight in this image. Photo by Dave Berman, 1993.

The team’s collective attitude changed when Stuart knocked off a chunk of bone-bearing rock from a bench in the quarry corner while shoveling away rock rubble. Careful examination of the fragment revealed part of the top of a roughly five-inch-long skull. We have since joked that Stuart gave it a lobotomy. While collecting the large block of rock containing the remainder of the skull, another piece of rock popped off the the edge of the block adjacent to the quarry wall. This piece had vertebrae in it. The team then realized that only the front portion of the animal was in the block freed from the quarry. The rest of the fossil skeleton remained in the quarry wall. Thomas later excavated the rest of the specimen and shipped it separately to Carnegie Museum of Natural History (CMNH).

You can watch Dave and Stuart excavate the fossil-bearing block by clicking on the video link at the end of this post.

Based on the shape of both the exposed teeth in the broken skull and the exposed vertebrae, Dave and Stuart were able to identify the fossil animal as the genus Diadectes. Thomas had already collected a juvenile skull and other bones of Diadectes before his collaboration with Dave, but the specimen discovered in 1993 was by far the most complete and best preserved.

Skeleton of Diadectes absitus. Some of the limb bones are preserved on the underside of the block. Photo provided by Thomas Martens.

Once my preparation of the specimen was completed, which took a little over a year, Dave, Stuart, and Thomas begin their detailed study and description of the fossil. They determined that it represented a new species, which they named Diadectes absitus. “Absitus” is Latin for distant or far, in reference to the species being the first occurrence of Diadectes outside of North America. The generic name Diadectes was coined in 1878 by the famous paleontologist Edward Drinker Cope, and is a combination of the Greek “dia,” meaning crosswise, and “dēktēs,” meaning biter, in reference to its broad teeth. Other species of Diadectes occur in similar-aged rocks in the American southwest, and a few specimens are known from the Tri-State area of Pennsylvania, Ohio, and West Virginia.

Diadectes is a member of the group Diadectomorpha, which has oscillated between being considered a member of Amphibia or Amniota. Amphibians lay their eggs in water, which then hatch into tadpoles that later undergo metamorphosis. Today this group includes frogs, salamanders, and caecilians (limbless, worm-like burrowing amphibians). In contrast, amniotes either lay their eggs on land, like reptiles and birds do today, or the embryo develops sufficiently in the mother for live birth, as in most mammals. Except in rare cases, the type of developmental pathways of fossil animals cannot be determined because they are rarely preserved with their eggs or fetuses. Paleontologists instead study a variety of preserved features to determine group membership. As an example, amphibians typically have four fingers, whereas amniotes generally possess five.

Diadectes and its close relatives were herbivorous, that is, they ate plants. Their spatulate, incisor-like front teeth project forwards and were adapted for cropping vegetation. Longitudinal, parallel striations on their broad cheek teeth suggest that Diadectes could move its lower jaw fore and aft to grind plant matter against its upper jaw teeth, a motion called propalinal.

Skull of Diadectes absitus in right lateral (= right side) aspect. Notice the forward-angled front teeth and the bulbous cheek teeth. A black pen was used to mark the boundaries of individual bones in the skull, which aided study of the animal. Modified from photo provided by Thomas Martens.

The presence of an enlarged torso and teeth adapted for grinding tough vegetation are evidence that Diadectes absitus likely consumed a diet of high-fiber plants. Animals that eat high fiber plants, such as cows, have enlarged torsos framed by a rounded rib cage to hold large guts for processing plant cellulose through fermentation by microorganisms.

Diadectes absitus lived at a time when herbivores were just beginning to evolve. One of the oldest known herbivores is the diadectomorph Desmatodon hollandi, which lived about 305 million years ago, whereas Diadectes absitus lived roughly 290 million years ago. We discovered a surprisingly high number of herbivores at the Bromacker.

Teeth of one of the oldest known herbivores, the diadectomorph Desmatodon hollandi. This specimen was discovered in Pitcairn, PA by Percy E. Raymund (Assistant Curator, Section of Invertebrate Paleontology) in 1907 and named in honor of Dr. William Holland, the second Director of CMNH. The teeth of Desmatodon are very similar to those of Diadectes absitus. Photo by the author, 2018.

A cast of the skeleton of Diadectes absitus is exhibited in the Fossil Frontiers display case in the Dinosaurs in Their Time exhibition. Be sure to look for these once the museum re-opens. And stay tuned for my next post, which features another diadectomorph, Orobates pabsti.

Photograph of a model of Diadectes absitus made by the Museum der Natur, Gotha exhibit preparator Peter Mildner. Photo provided by Thomas Martens.

Dave and Stuart excavate the fossil-bearing block (video)

Finding fossils at the Bromacker quarry was tedious and physically demanding, but it was extremely rewarding when a fossil is discovered. Our annual summer field season generally lasted three and a half weeks. Because the weather usually wasn’t conducive to camping or cooking outdoors, we stayed at the same hotel and dined at the hotel or local restaurants.

The original 1993 fossil quarry was opened using heavy equipment and operators from the nearby commercial quarry, and in the early years we relied on these people to expand the fossil quarry’s boundaries as needed. When the commercial quarry was temporarily shut down due to the lack of contracts for building stone, our collaborator Dr. Thomas Martens fortunately was able to obtain funding to annually rent a Bobcat, which he became skilled at operating. Thereafter, Thomas would use the Bobcat to expand the quarry and remove soil and weathered rock layers, so that we could begin our yearly excavation on unweathered rock.

The Bromacker quarry on the first day of fieldwork in the 2006 field season. Shovels were used to clear loose rock from the surface of the quarry. Pictured (counterclockwise from left) graduate student Andrej Čerňanský and Dr. Jozef Klembara (Comenius University, Bratislava, Slovak Republic) and Dr. Dave Berman (Carnegie Museum of Natural History [CMNH]). Photo by the author, 2006.

We would each stake out an area of the quarry to work in and then proceed to work through the rock layers by using a hammer and chisel or pry bar to free a piece of rock. Its surfaces and edges would be checked for fossil bone, and if there was none, the rock piece would be broken into smaller pieces, which were also checked for bone. As is the case at many other fossil sites, the rock tended to split along the plane a fossil was preserved in, because the fossil would create a zone of weakness.

The quarry after a couple days of excavation. Pictured (clockwise from front) are Dave Berman, Jozef Klembara, and Andrej Čerňanský. Photo by the author, 2006.

Once a fossil specimen was discovered—and there were a few frustrating years when this didn’t happen—the hard work of extracting it from the quarry began. Here, I’ll use a fossil discovered during the 2006 field season as an example of how this was done.

A discovery! Fossil bone and bone impression are exposed to the left of the lens cap. Photo by the author, 2006.

First, we would isolate the fossil specimen from the surrounding rock, exposing as little of the fossil as possible while determining its extent, because it would have been easy to lose pieces of bone in the dirt and mud. Then we would encase the specimen in a plaster and burlap jacket to protect it during extraction, shipping, and preparation.

To make the jacket, we’d coat cut strips of burlap in wet plaster and then spread them across the surfaces of the rock containing the specimen, or block. A layer of plastic (plastic bags worked well) was applied to the top surface of the block to keep plaster from sticking to any exposed bone.

The block is partially isolated from surrounding rock. In this case, we decided to encase the top and some sides of the block in plaster and burlap bandages to hold the rock pieces together before we finished isolating the block from surrounding rock. Photo by the author, 2006.

After a couple layers of plaster bandages were applied to the top and sides of the block, the block was undercut, with plaster bandages added periodically to hold the undercut rock in place.

Pictured (left to right) are Dr. Stuart Sumida (California State University, San Bernardino, CA), Dave Berman, and Mr. Jerome Gores (Museum der Natur, Gotha [MNG]). Jerome is holding a plaster and burlap bandage while Dave and Stuart are pressing plaster bandages against the bottom of the block. They must hold the bandages in place until the plaster sets. Photo by the author, 2006.

Using hammers and chisels to undercut the block. Photo by the author, 2006.

When deemed safe, we would crack the block free from the quarry floor using hammers and chisels, and flip it over, unless it broke free on its own. Excess rock would be removed from the bottom of the block to make it lighter in weight. Then we would apply burlap and plaster bandages to the bottom of the block. The block would be removed from the quarry and stored at the MNG until it was shipped to Pittsburgh.

The block has now been safely flipped over and excess rock is being removed. Photo by the author, 2006.

We encountered several problems during our quarry operations over the years. As we worked our way through the rock column in the quarry, processed rock piled up on the quarry floor. In the early years, we tossed or shoveled the processed rock into wheelbarrows and pushed the heavy, unwieldy wheelbarrows out of the quarry to a dump pile. Fortunately, the Bobcat eventually replaced the wheelbarrows for moving processed rock. As we ran out of space outside the quarry to dump processed rock, the rock was used to backfill older portions of the quarry.

Former CMNH volunteer Linda Rickets (front, right) and the author (left, rear) line up to push loaded wheelbarrows out of the quarry to the dump pile. Photo by Dave Berman, 1996.

At the dump pile. Over time rain and the freeze/thaw cycle would break down the rock and vegetation would grow on it. Photo by Dave Berman, 1996.

Another problem was that fossils found at the bottom of the quarry were often extremely difficult to undercut because the rock so was hard. Sometimes a well-hit chisel would just bounce off the rock instead of cracking or penetrating it. One year we had to resort to a rock saw to undercut a block.

Dave Berman uses a rock saw to undercut a block. Photo by the author, 2004.

Rain was always a problem. We would shelter in our cars during intervals of rain, or work at the museum if the rain was heavy and persistent. Occasionally heavy rain would flood the quarry, forcing us to work in the ‘dry’ areas of the quarry while a pump drained the water. Of course, we had contests to see who could skip a rock the farthest or make the biggest splash.

A flooded quarry greeted us in the morning after heavy overnight rain. Pictured are Thomas Martens (front) and Stuart Sumida (rear). Photo by the author, 2010.

Next week’s post will describe the process of fossil preparation, that is, removing rock to reveal a specimen in the lab. The fossil collected in the 2006 field season will be used as an example.

Here are some videos taken by taken by Thomas Martens’ wife, Steffi, during the 1993 and 2006 field seasons. These show the process of searching for fossils (1993 video) and collecting the fossil highlighted in this post (2006 videos).

Bromacker Quarry 1993

https://youtu.be/DAEG0l1NotE

Bromacker Quarry 2006

https://youtu.be/UNL1s5ycJSM

https://youtu.be/lqYxheOZLQY

https://youtu.be/HrmSBrhde_E

Two exquisitely preserved, nearly complete adult skeletons of Seymouria sanjuanensis that were discovered in the Bromacker quarry in 1997. Photo by Dave Berman.

At lunchtime on the last day of the 1997 field season, Thomas Martens discovered the two exquiste specimens shown above, the only fossils found that year. Thomas had uncovered a piece of the hip region with some attached vertebrae that resembled, once again, those of the ancient amphibian Seymouria. Because our work time was limited, we estimated the length of the specimen and rushed to extract it from the quarry. When we flipped the block over, a few pieces of rock fell out, revealing a series of vertebrae of a second individual in the block. We were thrilled to learn that Thomas had discovered two specimens of Seymouria. We put the rock pieces back in place and quickly finished plastering the block. There was just enough time for Dave, Stuart Sumida, and I to return to our hotel, clean up, quickly pack, and meet Thomas, his family, and his fossil preparator Georg Sommer for a celebratory dinner. What a great way to end the field season.

Working in tight quarters to quickly extract the Seymouria specimens discovered at lunchtime on the last day of the field season. Clockwise from right: Georg Sommer, Dave Berman, and the author. Photo by Stuart Sumida, 1997.

Seymouria had already been known from the Bromacker quarry. Thomas had discovered and identified two skulls in 1985, fossils he brought with him when he came to Carnegie Museum of Natural History (CMNH) in 1993 to study for six months with Dave Berman under a CMNH-financed fellowship. Both skulls were of juvenile individuals. Of the two known species of Seymouria, Dave and Thomas were excited to discover that the Bromacker skulls were nearly identical to those of Seymouria sanjuanensis. The 1997 lunchtime discovery of the two complete adult specimens confirmed the identification of the Bromacker Seymouria as S. sanjuanensis.

The first discovered species of Seymouria was Seymouria baylorensis, from near Seymour, Baylor County, Texas, from which its name was derived. Seymouria sanjuanensis was first found in San Juan County, Utah, by Dave Berman and the field team he was leading as a graduate student at the University of California, Los Angeles. Dave’s advisor, Dr. Peter Vaughn, named it Seymouria sanjuanensis in reference to the county of discovery. Another discovery of five specimens of this species preserved together was made by Dave in New Mexico in 1982.

Comparison of the skulls of Seymouria baylorensis (top) and S. sanjuanensis (bottom). The individual bones of the skull are color coded. Skulls scaled to same size. Image from Wikimedia Commons.

Seymouria baylorensis is geologically younger than S. sanjuanensis and has a more robust skull, larger and fewer teeth of variable size, and a subrectangular postorbital bone compared to the chevron-shaped postorbital of S. sanjuanensis.

Seymouria is considered a terrestrial amphibian that only returned to water to breed. Its strongly built skeleton provided the support needed to move on land. With its numerous, slender, pointed teeth, S. sanjuanensis most likely ate insects and small land-living vertebrates. We know that the Bromacker Seymouria didn’t consume fish, because not a single fish fossil, scrap of fish fossil, or fish coprolite (fossil poop) has ever been found at the Bromacker quarry. Study of the rock deposits preserving the fossils at the Bromacker indicate a lack of permanent water, which would explain the absence of fish.

Growth series of skulls of Seymouria sanjuanensis from the Bromacker Quarry showing (left to right) early juvenile, late juvenile, and adult growth stages. Photos by the author, 2006.

Conditions for breeding must have been favorable in the Tambach Basin, the ancient basin where sediments preserving the Bromacker fossils accumulated, because several juvenile specimens of Seymouria are known. The smallest is a skull measuring about 3/4 of an inch long. In a study led by our colleague Josef Klembara (Comenius University, Slovak Republic), we determined that the smallest individual was post-metamorphic—in other words, no longer a tadpole—based on the presence of certain ossified bones in the skull. In tadpoles, these skull elements are cartilaginous; that is, they haven’t yet turned to bone.

Five skeletons of Seymouria sanjuanensis preserved together were discovered in north central New Mexico by Dave Berman in 1982. These specimens are on display in CMNH’s Benedum Hall of Geology, in the “What is a Fossil?” case. Photo by the author, 2013.

The discovery in Germany of the same species of Seymouria previously known only from New Mexico and Utah has important implications in terms of paleobiogeography (the study of the distribution of species in space and time). At the time S. sanjuanensis was alive, the continents were merged to form the supercontinent Pangaea. The presence of S. sanjuanensis across Pangaea, north of a roughly east-west trending mountain range, indicates that climatic or physical barriers (e.g., deserts, inland seas, mountain ranges) didn’t prevent its dispersal.

Map showing the arrangements of the continents in the Early Permian. The locality where Seymouria occurs in present-day New Mexico, Texas, and Utah and the Bromacker locality in present-day Germany are indicated. Map modified from Scotese, 1987.

The two Seymouria specimens preserved together were a big hit in the local region in Germany. Museum der Natur (MNG) exhibit preparator Peter Mildner nicknamed them the “Tambacher Liebespaar” (“Tambach Lovers”) after a painting entitled “Gothaer Liebespaar” (“Gotha Lovers”) on exhibit in the Herzogliches Museum of the Stiftung Schloss Friedenstein (also the parent organization of MNG). This name caught on and is fondly used by our German friends and colleagues. Peter even made a fleshed-out model of the two Seymouria specimens in their death pose. The proprietor of the hotel in which we stayed hung a copy of the model of the Tambach Lovers and a framed collage of newspaper articles featuring the Bromacker on a wall in one of the hotel rooms, which she named the “Präparation Suite” (i.e. “Preparation Suite” in reference to the preparation of fossils). I often stayed in this room.

The painting entitled “Gothaer Liebespaar” (“Gotha Lovers”), which is on display at Herzogliches Museum of the Stiftung Schloss Friedenstein, Gotha, Germany. Image from Wikimedia Commons and provided by Thomas Martens.

Postcard showing the Tambach Lovers. The postcard was made for and sold by the Museum der Natur, Gotha. Photo of the postcard by the author, 2020.

Stuart Sumida (left) and Heike Scheffel, proprietor of the Hotel Wanderslaben where we stayed (right), with the model of the Tambach Lovers in the “Präparation Suite.” The framed collage to the right of the model holds newspaper articles featuring the Bromacker project. Photo by the author, 2003.

A cast of the Tambach Lovers specimen and a model of Seymouria sanjuanensis are exhibited in the Fossil Frontiers display case in CMNH’s Dinosaurs in Their Time exhibition. Be sure to look for them once the museum re-opens. And stay tuned for my next post, which will feature the unusual bipedal reptile Eudibamus cursoris.

For those of you who would like to learn more about Seymouria sanjuanensis, here is a link to the publication describing the 1997 specimens: https://www.tandfonline.com/doi/abs/10.1671/0272-4634(2000)020%5B0253%3AROSSSF%5D2.0.CO%3B2.

In 1995, my first year of field work at the Bromacker quarry, Stuart Sumida discovered a fossil that we initially thought was that of the amphibian Seymouria, based on the size and shape of the exposed vertebrae. This tentative identification made sense, because before our collaboration began, Thomas Martens had discovered in the Bromacker a skull of Seymouria, a creature known from localities in the USA. Months later, while I was preparing the specimen, Dave Berman and I realized the fossil wasn’t Seymouria, and that it belonged to the same unnamed animal that Thomas had collected a partial skeleton of before our collaboration began.

Specimen of Orobates pabsti collected in the 1995 field season. We determined that it is a juvenile. Photo by Dave Berman.

In the 1998 field season I discovered a third specimen, which is by far the most spectacular fossil that I have ever discovered. I found it towards the end of the field season when I pried up a piece of rock from the quarry floor. Upon turning over the rock piece, I saw an articulated foot preserved in it. I couldn’t believe my eyes! I knew that at the Bromacker if an articulated foot was found, the rest of the articulated skeleton should be attached to it. The problem was, we didn’t know if I had discovered a front or a hind foot, so we weren’t sure how the specimen was oriented in the quarry and whether it penetrated the nearby rock wall. Dave carefully lifted another piece of rock and thought the bones exposed in it were part of the shoulder girdle. Unfortunately, closer examination revealed that it was a piece of skull roof—another lobotomy—but, lacking x-ray vision, this is how we find fossil bone at the Bromacker. The good news was that the fossil specimen appeared to parallel the quarry wall.

A film crew from the regional MDR television station visited us early on the day of my discovery to interview Dave and Thomas. The discovery was made after they left, so Thomas immediately notified them. They returned and recorded a reenactment of my discovery. The piece of rock I am holding contains the foot. The rest of the fossil lies in the low mound of rocks in front of me. Photo by Dave Berman, 1998.

Dave and Stuart finish plastering the block. The red flag is a north arrow to indicate the orientation of the block in the quarry. Photo by the author, 1998.

Dave, Stuart, Thomas, then-graduate student Richard Kissel (University of Toronto, Mississauga), and I named the animal Orobates pabsti, which is from the Greek “oros,” meaning mountain, and “bates,” meaning walker, in reference to the Bromacker fossil environment being an intermontane basin. “Pabsti” is in honor of Professor Wilhelm Pabst for his pioneering work on the Bromacker fossil trackways.

We determined that Orobates is very closely related to Diadectes, and like Diadectes, was herbivorous. Orobates differs from Diadectes and other diadectomorphs in the group Diadectidae in a number of features, some of which are as follows: spade-shaped cheek teeth that are oriented on the jaw at an angle of 30–40° to the jaw line, rather than being close to 90°; narrower and shorter vertebral spines; 26 vertebrae between the head and hip (Diadectes has 21); proportions and shapes of individual toe bones; and digit (finger or toe) length.

Holotype specimen of Orobates pabsti, the specimen collected in 1998. If a series of specimens exists of a new species, then the specimen that best represents the species is designated as the holotype. If only one specimen is known, it becomes the holotype by default. Photo by Dave Berman.

The Bromacker has long been famous for its exquisitely preserved fossil trackways. Identification of the particular fossil animal that made a given trackway is almost always very difficult, because body fossils often lack completely preserved hands and feet and typically are not found in association with trackways. As a result, trackways are given their own set of names, called ichnotaxa (“ichno” means track or footprint), which are typically referred to major groups of animals instead of individual species. The Bromacker is unique, however, because nearly completely preserved body fossils occur in a rock unit above the trackways, indicating they are very nearly contemporaneous. Five ichnotaxa are known from the Bromacker, and one of them, Ichniotherium, has been attributed to Diadectidae.

A large slab of rock being inspected for trackways shortly after it was unearthed in the commercial rock quarry. The polygonal patterns in the rock are mudcracks. Photo by the author.

Graduate student and trackway expert Sebastian Voigt (now Director at Urweltmuseum GEOSKOP, Burg Lichtenberg, Germany) often visited us at the Bromacker. In 2000, a time when Diadectes was the only known Bromacker diadectid, Sebastian and his advisor Hartmut Haubold (now emeritus at Martin Luther University Halle-Wittenberg, Germany) proposed that Ichniotherium cottae made two track types, designated as A and B, that differed according to the speed at which the trackmakers moved. This contrasted previous studies that proposed three species of Ichniotherium at the Bromacker.

Once the skeletal anatomy of Orobates became known, Sebastian realized that there were two species of Ichniotherium, and they were made by Diadectes and Orobates, respectively. He invited Dave and me to co-author a paper to present this hypothesis. We supplied Sebastian with information about skeletal differences between Diadectes and Orobates, and Sebastian used these data to firmly establish that Diadectes made Ichniotherium cottae (type B) tracks and Orobates was the trackmaker of trackways formerly identified as I.sphaerodactylum (aka I. cottae type A). Even though the makers of the trackways are now known, the ichnotaxon names are still used when referring to the trackways.

Photographs of trackways of Ichniotherium sphaerodactylum made by Orobates pabsti (top) and Ichniotherium cottae made by Diadectes absitus (bottom). Modified from Voigt (2007).

In Diadectes, the fifth digit of the hind foot is relatively shorter than it is in Orobates, which can be seen in the tracks of I. cottae and I. sphaerodactylum, respectively. Furthermore, in I. cottae trackways, the hind foot track overlaps the track of the front foot, whereas in I. sphaerodactylum the hind foot track typically doesn’t overlap the front foot track. This is because Diadectes has less vertebrae between the head and hip (21 vertebrae) than Orobates (26 vertebrae) does.

Front and hind foot track pair of Ichniotherium sphaerodactylum. Track made by the front foot is above the hind foot track. Digits 1–5 indicated. Modified from Voigt, 2007.

Front and hind foot track pair of Ichniotherium cottae. Track made by the front foot is above the hind foot track. Digits 1–5 indicated. Notice that the hind foot track overlaps the front foot track. Drawings are of different specimens than the one photographed. Modified from Voigt, 2007.

A cast of the holotype skeleton of Orobates pabsti is exhibited in the Fossil Frontiers display case in Carnegie Museum of Natural History’s Dinosaurs in Their Time exhibition. Be sure to look for it once the museum re-opens. And stay tuned for my next post, which will feature the amphibian Seymouria sanjuanensis.

For those of you who would like to learn more about Orobates, you can access the abstract here or contact Amy Henrici here. The publication on the track-trackmaker association can be found here.