One of the many mysteries we have about the sloth site is why we have so many more bones from the right side than left side of the adult.  (The photo (below) is misleading—it comes from an outreach program where we had to spread the bones out so that viewers on the “bad” side had something to look at.  The real difference isn’t absolute, but it’s striking.) There’s no intrinsic reason why the bones on one side of the sloth should have fossilized better than the other, so it must indicate something about the conditions near the time of death. (photo borrowed from)

Gary Haynes reports that when African elephants die, if they don’t drop in their tracks, most lie down on their left sides (Haynes, 1988).  It’s only a “strong . . . impression,” though, and he hasn’t a clue why it should be. Bones on the “down” side, have a better chance of being covered by water or sediment and protected from scavengers under some circumstances, and they may be more likely then to survive and tell the story of their early taphonomic history.  Haynes found a mammoth scattered at a site in Alaska where the transverse processes of the thoracic vertebrae were all heavily gnawed just on their right sides, indicating that the animal fell exactly as predicted (Haynes, 1980).

Different microenvironments that become established inside a carcass during decomposition could also be a factor. The role of the decay and putrefaction microbes in the process of fossilization is much debated, but there is no doubt that a mega-sized mammal like a sloth creates its own internal environment as it decomposes, with sufficient volume, mass and thermal inertia for the process to proceed for some time independently of external conditions.   That’s especially true in the winter without the insects to pierce the skin and interrupt the anaerobic bacteria. Lying on the ground, the “down” side of an animal as big as an adult sloth, removed from the desiccating effects of the sun and wind, or the cold, will stay warm and humid, creating a microhabitat for different microbes to thrive, selectively fostering or hindering the process of fossilization (Coe, 1978).

Presumably anything that put the sloth on its side, and kept it there, would produce the same effect.  But, any protection offered by dying on one side versus the other is apt to be fleeting as carnivores will usually roll a carcass over to gain access to the parts underneath.  However, large carcasses may be difficult for scavengers to turn.  Frison and Todd report it took 15 people pulling on ropes to flip the carcass of a large male African elephant they were butchering, even with the top side defleshed (Frison and Todd, 1986).  

Turning over a 1-ton sloth could be difficult too, depending on the scavenger, and impossible in the winter, with one side frozen to the ground.   Winter is the time of bounty for most carnivores. Hunger, stress and the snow leave prey vulnerable and relatively plentiful. Under normal circumstances predators can afford to ignore a frozen carcass and go hunting for something still warm and breathing. Consequently, winter carcasses are usually far less utilized than those of the summer dead or killed (Haynes, 1980).   Winter weather, access to only half of the adult sloth, and an abundance of other food, may go a long way toward explaining the condition of our adult’s bones and the contrast with those of the juveniles. . . but, is there any way to determine if  the sloths died in the winter? . . . Dave

References

Coe, M. 1978. The decomposition of elephant carcasses in the Tsavo (East) National Park, Kenya.  Journal of Arid Environments 1: 71-86.

Frison, GC and Todd, LC. 1986. The Colby mammoth site:  taphonomy and archaeology of a Clovis kill site in northern Wyoming.  University of New Mexico Press, Albuquerque, 238 pp.

Haynes, G. 1980. Evidence of carnivore gnawing on Pleistocene and recent mammalian bones. Paleobiology 6: 341-351.

Haynes, G. 1988. Longitudinal Studies of African elephant death and bone deposits.  Journal of Archaeological Science 15: 131-157.

According to Haynes (1988) every wolf pack has its hierarchy and once the prey is dead, and often even before, the process of staking out claims on preferred cuts and dividing the carcass begins. The dominant wolf will take the choice position. The blood and internal organs inside the abdomen are a favorite first pick. Sternal elements and ribs are usually damaged in the process. Other high ranking wolves will claim the rump and upper legs where there are large masses of flesh. If there are more animals in the pack than can comfortably situate themselves around the carcass to eat without invading each other’s space, they’ll start disarticulating the limbs, causing distinctive damage on the proximal ends of the femora and humeri , and their anchoring points on the pelvis and scapulae. The prizes will be carried a short distance away to be gnawed on in relative solitude. Lower ranking wolves will tear off smaller less desirable parts (e.g. ears, tail, jaw/tongue) and carry them further away. These satellite consumption spots will be randomly distributed around the carcass and about 20 feet apart (Haynes, ibid.). Tooth marks and distance will be correlated as wolves lower in the pecking order tend to invest more time gnawing on their meager rations instead of trying to muscle in and steal another portion.

The scatter of bones at a wolf kill site rarely exceeds 100 ft. from ground zero. Within this area one can expect to find the skull, ribs and vertebrae, but rarely the sternal elements, patelae (knee bones) and caudal (tail) vertebrae. A few of the major limb bones are likely to be missing—more if scavenging has been heavy. The lower legs of hoofed animals offer more tendons and ligaments than muscle so they are often discarded uneaten and the bones are often found in anatomical sequence. If scavenging is heavy, the lower leg may represent a prized morsel and elements may be transported miles from the kill site (Haynes, 1985). The scatter is normally lower at a non-kill/scavenging site, though that will increase over time with scavenging by a host of creatures and kicking. Most, if not all of the major limb bones should be close by. Wolves will normally leave even a fresh-dead prey carcass relatively intact. Unless they are particularly hungry, wolves simply prefer to do their own “shopping” (Haynes, 1982).

Wolves follow predictable sequences in disarticulating and consuming a kill. Their pattern of utilization varies with the prey species and its particular anatomical characteristics, its size, how hungry the wolves are, how many individuals are in the pack, and the season. The chart below summarizes his findings for large prey over 300 kg (~660 lbs.–e.g. moose or bison size). According to Haynes, the patterns are so regular for a particular predator and prey species that out-of-sequence disarticulation or damage, or the absence of a normal step, are reliable indicators of scavenger activity instead of predation, or other disturbance processes such as trampling, etc. (Haynes, 1982).

Carnivore Utilization Stages
	Stage I: light to moderate	Stage II: Full	Stage III: Heavy
Ribs	sternal ends consumed	Many on one side broken off below their articulating ends and scattered	further broken up
Femur	greater trochanter damaged & trochlear rim scored	disarticulated from pelvis, distal condyles gouged. Toothmarks undercut head. shaft lightly scored	head removed, distal end gone, shaft broken
Tibia		still articulated to femur	disarticulated, proximal end gone
Humerus	head and greater tuberosity furrowed and gouged	disarticulated from scapulae, greater tuberosity gone, shaft lightly scored	proximal end gnawed off, approx. 1/3 of proximal shaft gone.
Ulna	olecranon process damaged
Pelvis	edges of ilia and ischia gnawed		gnawed down to acetabula
Scapula	vertebral border damaged, still attached to humerus.	disarticulated from humerus, edge splintered	gnawed down to glenoid process
vertebrae		spines and lateral processes gnawed or broken off	Only a few badly gnawed vertebrae remain articulated
Skull	Still articulated to the body, no damage to bones. Nasal cartilage gnawed	disarticulated Nasal bones gnawed	Only toothrows remain from head
Mandibles	partly defleshed, articulated	defleshed.	Mandibles disarticulated

adapted from Haynes1982 and 1999

It’s probably too soon to use scatter diagrams and limb bone tallies to test the predator theory with the sloths. We still have the entire south bank of the creek to explore and map. However, we do have a large sample of bones to appraise using Haynes’s rules of carcass utilization. The ribs we have range from perfect to stage III. Our sole femur (R), the truest indicator of predation according Haynes (1982), shows signs of light scavenging by small mammals but none of the damage to the greater trochanter that he predicts from disarticulation by predators. The other femur (L) is missing entirely. Both tibiae are missing as well. Humeri heads are toothmarked and overall the bone shows Stage III damage, as do the vertebrae and the pelvis. The one ulna that survived (R) looks untouched—no signs of the disarticulation damage at the olecranon process (elbow) that he predicts in the case of a kill. One scapula (R) is almost pristine, but the other one is heavily damaged by trampling–no evidence of gnawing however. The skull is stage II while the mandibles are stage III.

Conclusion: no traces of disarticulation by predators judging from the adult bones we’ve found so far. However, most of the “baby’s” bones are still missing. All of the “toddlers” major bones are AWOL too, except for the distal half of one humerus. If we assume the baby weighed about 90# and the toddler perhaps 300# (based on scapulae, the only common bone we have), that’s about 200# of flesh–a healthy dinner for a pack of wolves. How much would a pack of wolves eat at one sitting?  Would they have turned their attention to the tender youngsters before eating a tough old adult? Would they have abandoned the adult’s carcass only half eaten? What accounts for the disparity in utilization stages of the carcass and why are the bones from the right side better preserved than those from the left? Hayne’s rules would say even if predation is confirmed, the deviations from the expected pattern indicate there are other factors at work. . .  but he also warns that the rules may be entirely different for ground sloths. As ever, we have more questions than answers, but Haynes’s research holds out the promise that with some more data maybe we can start to peel back the layers. . . . Dave

References

Haynes, G. 1982. Utilization and skeletal disturbances of North America prey carcasses. Arctic 35: 266-281.

Haynes, G. 1985. On watering holes, mineral licks, death, and predation. In Environments and Extinctions: Man in late glacial North America, Eds. J Mead and D Mettzer. Center for the Study of Early Man.

Haynes G. 1988. Prey bones and predators: potential information from analysis of bone sites. Ossa: 7: 75-97.

Haynes, G. 1999. The role of mammoths in rapid Clovis dispersal. In Mammoth and mammoth fauna: studies on an extinct ecosystem. Proceedings of the first International Mammoth Conference, St. Petersburg, Russsia. P. 9-38.

The Ice Ages offered a formidable cast of killers, but attacking a healthy adult Megalonyx, (a.k.a. “Giant Claw”) wasn’t a job for a solitary predator. Downing a giant sloth demanded teamwork. The giant short-faced bear (Arctodus simus) was once thought the greatest terror on four legs and maybe capable of going it alone, but modern studies suggest its size owed more to its herbivorous habits and bone-crushing/scavenging abilities, and at best it was only an opportunistic predator (Emslie and Czaplewski, 1985). The number of sabertooth cats (Smilodon fatalis) found at La Brea with serious wounds suggests a social structure that made it possible for them to survive even crippling injuries, but whether that cooperation extended to hunting is TBD. Evidence from Friesenhahn Cave in Texas indicates it was used as a den by the more lightly-built sabertooth, Homotherium serum, the scimitar cat. All the living Felids that den are solitary, however (Rawn-Schatzinger, 1992). The remains in the cave suggest H. serum specialized in racing in under the noses of momma mammoths and dispatching careless two year-olds with a strategic bite or two, and then dashing away apparently to wait for the youngster to bleed to death and the herd to depart, no doubt in frustrated fury. Other Homotherium species in the Old World followed similar practices with juvenile mastodons and rhinos (Lange, 2002). Bits of Harlan’s sloth (Paramylodon harlani) have been recovered from Friesenhahn Cave but there’s no evidence a cat brought it there, much less killed it (Graham, 2007). The scimitar cat’s style of hunting makes it a solid candidate for causing the wound we’ve found on our toddler’s back, but it would not have tried the same trick on an adult sloth. American lions (Panthera leo atrox) probably hunted like their African cousins, by stealth and ambush around watering holes and other locations frequented by their prey (Grayson, 1991). The absence of any fossils from eastern North America suggests, like their modern descendents, they preferred open habitat to the woodlands favored by Megalonyx (McDonald and Anderson, 1991). The lifestyle of dire wolves (Canis dirus) is still debated, but their massive teeth suggest scavenging probably played the preeminent role in their diet. If predators killed our sloths, the prime suspect has to be a social hunter like the timber wolf (Canis lupus). That’s lucky, because Gary Haynes has spent a lifetime studying wolf kill sites.

Every predator has a unique approach to eating a specific prey species and even when they leave no distinctive tooth marks (which is the rule), they often leave a telling signature in the specific areas they damage bones, the elements they ignore and the arrangement of the carcass when they leave. When wolves prey on moose and bison the most reliable indicator is the damage to the femora (Haynes, 1980b). Even with very light feeding, wolves destroy the greater trochanter. Now the greater trochanter of any prey species is hardly a tasty tidbit–wolves that attack here have one purpose—cutting the muscles that hold the bone in the acetabulum or hip socket. Predators are dangerous–to other animals and sometimes to each other. Even social predators can turn decidedly asocial in the blood lust of a kill—they demand elbow room. The first principle for recognizing the kill site of a social predator likethe timber wolf is looking for the evidence of the divvying up the spoils and creating the needed separation.

Next week—a look at our sloth femur: Dancing with wolves . . . . Dave

References

Did predators killed the sloths?  Last week Holmes and I were looking at a rib from the adult that’s currently  on display in the lobby when we noticed a large puncture and some adjacent gnaw marks. The wounds are partially obstructed by the case and easy to overlook.  They were obviously caused by  large sharp teeth and indicate a carnivore was present close to the time of death.  Carnivores don’t gnaw on bones to sharpen their teeth like rodents.  They may mouth an old dry bone they happen across, but nothing more.  If a carnivore bit into this bone, there was meat on it.

Identifying the forces that  can break and shape bones has been a concern of archaeologists for decades as intriguing assemblages of bones and bone fragments have surfaced at sites like Africa’s famous Olduvai Gorge.  The question was whether the remains indicated human hunting, and if the  bone splinters and spiral fractures were evidence of tool manufacturing.  Paleontologists like Voorhies (1969), Behrensmeyer (1975) and Hill (1976) broke many hearts showing stream transport, scavenging, weathering or trampling were responsible more often than humans. 

Predators don’t always leave obvious calling cards.  Haynes reports often finding  wolf kill sites with nary a scratch on the bones (Haynes, 1980).  Behrensmeyer (1975) and Voorhies (1969) showed a pattern of survival of the ends of the major limb bones was a strong indicator of carnivore activity. Predators and scavengers normally disarticulate limbs at their proximal or ball-joint ends first, damaging the bones at predictable points of muscle attachment.  The proximal ends of the humerus and femur also have very thin walls and offer an easily accessible and delectable center of marrow.   Few carnivores will pass it up unless they are overwhelmed with meat such as at the scene of a mass-drowning or other catastrophe.  Carnivores usually ignore the distal ends of humeri and femora (i.e.  elbows and knees).  Those joints are tightly wrapped in tendons and ligaments and offer little in return, so they are more likely to survive untouched and become fossilized.  Is it just a coincidence that we have three nearly identical distal ends of humeri (2-adult, 1-toddler)?

But fingerprints at the scene don’t prove their owner caused the deaths.  Carnivores are as likely gnaw off the end of a humerus or puncture a rib stripping the flesh from the steaming carcass of a fresh kill as cold carrion. How are we to determine if our sloths spent their last moments in a terrifying struggle against a pack  of predators, or simply died a “natural” death (e.g. from disease, malnutrition, drowning, etc.), and their remains were simply scavenged post-mortem?

Gary Haynes has spent a lifetime studying predator kill sites, and the patterns he has found could add an exciting new dimension to our understanding of the sloth site.   Resources, according to Haynes,  are distributed in predictable ways in the environment and to survive  animals learn the patterns all around them and across the seasons. Predators survive by mirroring that environment, especially the behavior of their prey,  and so their behavior becomes extremely patterned too, including at the site of a kill.  The location, distribution, condition, and type of bones or bone fragments  left at a kill site–sometimes even a single bone, can be distinctive enough to distinguish predator activity from scavenging and the presence  of a particular carnivore–no punctures or gnaw-marks required!  A wealth of information about an ecosystem may be revealed, such as how many wolves, for example,  were in a pack, the season of the year when they made the kill, how hungry they were, how vulnerable their prey was . . . even their favorite NFL player.  If Ice Age predators behaved like their modern-day cousins and followed the same general ecological rules–and Haynes believes  they did–relatively undisturbed bone assemblages such as ours can reveal much about predator-prey interactions, even involving extinct species (Haynes, 1982). . . . Dave  (to be continued)

References

Behrensmeyer, AK. 1975.  The Taphonomy and Paleoecology of Plio-Pleistocene Vertebrate Assemblages East of Lake Randolph, Kenya, Bulletin of the Museum of Comparative Zoology 146: 473-578.

Haynes G. 1980. Prey bones and predators:  potential information from analysis of bone sites.  Ossa: 7: 75-97.

Haynes, G. 1982. Utilization and skeletal disturbances of North America prey carcasses.  Arctic 35: 266-281.

Hill. AP. 1976. On carnivore and weathering damage to bone. Current Anthropology 17:335-336.

Voorhies, MR. 1969.  Taphonomy and population dynamic of an early Pliocene vertebrate fauna, Knox County, Nebraska, University of Wyoming, Contributions to Geology, Special Paper No. 1. 

External conditions (i.e. soil chemistry, microbes, etc.) are generally believed to determine the fate of bones, but Bell et al. (1996) suggest internal factors are more important. The bodies of all animals contain a host of microorganisms in their guts ready to escape after the death of their host and use its vascular and lymphatic systems to invade the major body organs. The microbes could be getting into the bone marrow by the Haversian canals that maintain the bones in life.  The actions of predators,  and their manner of killing–specifically, disemboweling prey and disrupting the microbial escape routes may be critical for explaining bone preservation.  I love the irony–fall to predators but live forever as a fossil.  Cool.

If you want to become a fossil, getting eaten by wolves may be your ticket.  Avoid hyenas–they’ll eat you bones and all. Lions and tigers will do a lot of damage to your bones too. A pack of wolves is perfect.  Haynes (1988) says they view internal organs as among the choicest bits and consume them first–bye bye gut microbes.  Better yet,  wolves rarely leave any tooth marks on the bones of a kill–hello immortality.

 Shipman (1981) says predation is overemphasized as a contributor to the fossil record, accounting for less than10% of the mortality  of the average prey species,  but we have evidence that our toddler survived at least one attack. Maybe our sloths weren’t so lucky the last time.  Micozzi (1986) adds some intriguing support. 

Forensic scientists have long studied the process of decomposition to estimate the time of death.  The test carcasses are often temporarily stored in a freezer.  Micozzi wondered if freezing and thawing were affecting the results and compared the decomposition of fresh rats to ones frozen and thawed.  He  found freezing killed the gut microbes that normally started the process of putrefaction.  Decomposition proceeded from the outside in, driven mostly by aerobic decay (i.e. oxygen-using microbes).  The fresh, non-frozen rats decomposed from the inside-out, driven by the internal anaerobic bacteria (living without oxygen).  Micozzi couldn’t follow his experiment long enough to determine if the defrosted rats were more likely to be fossilized, but clearly they were on a different chemical/microbial pathway.

Trueman and Martill (2002) suggest gut bacteria may be the initial attackers of bone–pioneers opening virgin bone to settlement by waves of followers, and anything that disrupted them–e.g. gutting by a predator, butchering or freezing halts the process of bone decay.   Once started, they believe decay proceeds to complete destruction.  Only those bones that escape attack by internal bacteria, or have them halted in their tracks by a rapid physical or chemical change, survive to become fossils.

So how can we tell if predators killed our sloths?  . . .    Dave

References

Bell, LS, Skinner, MF, Jones, SJ . 1996. The speed of post mortem change to the human skeleton and its taphonomic significance, Forensic Science International 82: 129-140.

Gill-King, H. 1997. Chemical and ultrastructural aspects of decomposition.  In Forensic Taphonomy: The Postmortem fate of human remains, WD Haglund and MH Sorg eds. pp. 93-108.

Haynes G. 1988. Prey bones and predators:  potential information from analysis of bone sites.  Ossa: 7: 75-97.

Micozzi, MS. 1986.  Experimental study of postmortem change under field conditions:  effects of freezing, thawing, and mechanical injury Journal of Forensic Sciences 31: 953-961.

Shipman, P. 1981. Life History of a Fossil:  An Introduction to Taphonomy and Paleoecology, Harvard University Press.

Trueman, CN and DM Martill, DM. 2002. The long-term survival of bone:  The role of bioerosion. Archaeometry 44: 371-382.

Can you explain the deaths of three sloths by anything other than some kind of catastrophe?  In a drought herbivores congregate around the remaining water sources, and soon exhaust all the good forage near by.  If the drought continues, eventually they die of malnutrition. Predators have plenty of meat available so they leave most of the carcasses undisturbed.  Even scavengers that normally consume bones switch to soft tissue.

Shipman (1975) identifies three levels of drought:  mild, when seasonal watering holes dry up and juveniles tend to die; severe when even permanent water sources are affected and animals in their prime die, and; extreme drought when entire lakes and rivers dry up and the environment is permanently altered.    If we had a drought, it was severe given the presence of an adult. We’ve found some turtle and frog bones, and mollusks, but there’s no evidence of an extreme die-off. The end of the Ice Age isn’t envisioned as drought prone, but we don’t have a firm radiocarbon date yet either.

Shipman offers some clues for recognizing a severe drought in a bone assemblage:  Often some of the bones are still articulated.  That’s because when the rain returns the dry soil erodes easily and covers them. The assemblage often includes a variety of animals from disparate habitats, forced together in their need for water. Also, the mineralogy of the sediments may show evidence of the arid conditions (i.e. hardpan).

Our bones show few signs of scavenging but they are disarticulated  (How do you explain the bone arrangement?). There’s no evidence of a pulse of sediments covering the bones (no layers evident anywhere in the clay).   Mineralogical analysis is awaiting the next NSF grant.  We haven’t found a single bone from another herbivore–strange if water was scarce, but would the presence of two juvenile sloths  make  “Mom” aggressive enough to drive other animals away?  The jury is still out on the question of drought—we need more data.

Haynes (1985) has studied elephant bone assemblages in Africa and notes that bones tend to accumulate near water for reasons other than catastrophic droughts: 1) animals spend more time there so they will die there more often simply from natural causes; 2) predators know that’s often the easiest place to make a kill; 3)  catastrophic floods may be more frequent there, and;  4) bones are more apt to be buried and preserved there.  We know there wasn’t a flood, but his other points merit some thought.

Conybeare and Haynes (1984) studied an assemblage of elephant bones after a severe drought at a place called Shabi Shabi in Zimbabwe. The assemblage  seemed to have a natural age distribution reflective of one catastrophic event, but they concluded it had actually accumulated over several years and that trampling, kicking and digging by the elephants, combined with compaction and drying had preserved  the assemblage as a single homogeneous bone bed.    The lesson is clear—don’t jump to any conclusions that our sloths died in a single event.  We have to prove it. . . . Dave

References:

Conybeare A. and Haynes, G.1984. Observations on elephant mortality and bones in water holes.  Quaternary Research 22: 189-200.

Haynes G. 1985   On watering holes, mineral licks, death, and predation.  In Environments and Extinctions:  Man in late glacial North America,  Eds.  J Mead and D Mettzer.  Center for the Study of Early Man. pp. 53-71.

Shipman, P. 1975. Implications of drought for vertebrate fossil assemblages, Nature 257:667-668. 

Heaven knows we’ve almost lost some students and one Bobcat operator to the muck at the site, but how reasonable is it to think a giant sloth could die that way? How about three sloths? Haynes (1988) has studied hundreds of elephant deaths and reports it’s actually not an uncommon event.   Healthy adult elephants never have a problem even in the deepest thickest mud but very young animals and those who are ill or weak some times get stuck.  He has also observed impala, Cape buffalo and black rhinos dying in this manner. 

So death by mud isn’t out of the question (Meghann take note).   But trapped animals leave some evidence behind after they are dead.  We would expect to find foot and leg bones locked in the sediments, articulated in a vertical or standing position.  We haven’t found anything like that.  Of course, the sediment could have become saturated with water again and then been disturbed by trampling and other activities at the water hole.  But that’s a lot of churning and the toddler, at least what we have found, is fairly localized.  A current strong enough to stir up the muck and clear it away would have moved a lot of bone too and left heavy sediment behind.   No, all the evidence says little or no current. . . . Dave

References

Haynes G. 1988. Longitudinal studies of African elephant death and bone deposits.  Journal of Archaeological Science 15: 131-157.

Why was it that only certain bones, and not others, got separated from the main concentration, and how did these end up here?  If we understood that, would it help us find the missing bones?

Under normal conditions bones don’t start moving until they are completely disarticulated (Hill, 1975).  A lot of researchers have tracked the decomposition of individual carcasses over time, but Hill picked out a large area in East Africa to study and recorded the status of disarticulation for every Topi, Damaliscus korrigum, a common medium-sized antelope, in his 475,000 square mile study area.  He found a surprisingly consistent pattern.

Under normal conditions, a mammal will follow the same sequence of disarticulation, regardless of how it dies. All vertebrates share the same general body form, so the sequence is basically the same way you carve a Thanksgiving turkey–head and tail, arms and legs, and finally the trunk.   Mobility demands a certain amount of loose attachment and after you’re dead everyone looking for a meal, be they a cannibal or carrion beetle; lion or wolf; maggot or mite; bacterium or fungus, follows the path of least resistance when they start dismantling you.  That’s determined by the architecture of your joints (i.e. tendons, ligaments, etc.).  Juveniles are different–their bones generally come apart at the epiphyses, where they are still growing and unfused, before they detach at the actual joints. The loosest (adult) connections are where your jaw attaches to your skull, your head attaches to your neck and where your tail (if you have one) and limbs attach to your trunk.  Hill, by observing thousands of skeletons learned the disarticulation sequence of each major unit or appendage.

Topi disarticulate: 1. scapula 2. caudal vertebrae 3. arm (from the scapula), 4. mandible (lower jaw) 5. hand and wrist (from the arm–the order of arm bones is somewhat variable) 6. skull and atlas 7. wrist (from the hand) 8. fingers, also leg 9-12. femur, tibia, tarsus, ankle, etc. 13. atlas (from skull–may go with axis in some animals) 14. toes 15. ribs (this is an average, some will go sooner, others later) 16-18. pelvis 19-21. vertebrae (Hill, 1979)  The details may vary for different mammals, but the general pattern probably applies to all, including ground sloths. 

The pattern of disarticulation may reveal some important information about a bone assemblage:

1) Significant variations may indicate the influence of an important environmental factor.  Water, for example, accelerates disarticulation and may change the sequence.  On land, animals tend to disarticulate from the trunk out (proximal to distal).  In the water the extremities of animals disarticulate more quickly (i.e. distal to proximal), though jaw and head are still in first place (Schafer, 1972).    Extremely arid conditions may dry tendons and ligaments around the ankle and wrist assemblies of smaller animals and inhibit disarticulation.  But that’s not likely to be an issue in our case, even if there was a drought.  Large animals, with their smaller surface area: volume ratio retain a lot of body water and bacterial putrefaction is likely to disarticulate the bones before a carcass mummifies.

2) Flowing water only transports individual bones, unless there’s a flood, so the disarticulation sequence determines what gets transported and scattered, and in what order.  Predators may move a group of bones initially but for the most part bones are dispersed individually.  Shouldn’t this be true of kicking as well?  

3) Disarticulation means some bones will be covered and protected by sediments sooner, and preserved. The survival of a fragile bone (like our scapulae?), among others less well preserved, may be explained by this.  Is it significant that most of the missing sloth bones come from the right side?

4) Disarticulation provides an estimate of the interval between death and burial by observing the degree of weathering on various bones.   The closest we’ve come to finding anything that might have still been articulated when it was buried is the string of 3 thoracic vertebra next to the pelvis.  Since they are among the last bones to disarticulate, the implication is the sloth had plenty of time to disarticulate and scatter before burial.

5) Disarticulation provides clues about which bones should be associated.  That can be important when they start being scattered.

Scattering

The scattering stopwatch starts with disarticulation of the first bone.  Hill observed that the largest concentration of bone is generally where the animal died.  If scattering is random and operates continuously, then the distance from the center, or “ground zero,” will correspond to a bone’s order of release as a single element–i.e. all other things being equal, the bones farthest from the center disarticulated the soonest.

Kicking can cause a significant amount of movement over time.  Bones are more likely to be kicked further apart than closer together, but the curve flattens over time.  At some point of dispersion, it’s almost as likely that a kick will propel a bone back closer to the center, as away, i.e. the dispersion field isn’t infinite and eventually stabilizes. 

 Illustration of differences in probability of dispersion at different distances of separation.  Bone B and bone B’ both move K units in a random direction in a single event.  B and D are K units apart, and B’ is 2K units from D’. The probabilities of B moving nearer to D, and B’ moving nearer to D’ are proportional to angle ABC and angle A’B'C’.  The general formula for this probability P=1-(arc cos (1/2r)/180) where rK is the distance separating the bones. (Hill, 1979)  

This means we don’t have an infinite area to search and maybe we can find a mathematics/statistics student who can make some assumptions about the average distance bones are kicked based on where we’ve found certain bones and calculate some probabilities for finding our missing bones within specific circles=r.  

There are still some unknowns:  1) The scapulae may disarticulate first but they don’t offer much vertical area for kicking.  Are they really likely to have been kicked the furthest? 2) Will animals avoid trampling and/or kicking especially large bones? Heavy bones?  Does it depend on the trampler/kicker (i.e. bison vs. mastodon)?   The mandible, atlas and caudal vertebrae seem the logical bones to focus on–everyone agrees they disarticulate early, regardless of conditions and they aren’t so large that most animals would avoid them.   But it still seems like a problem someone could model on a computer. . . .  Dave

References

Hill, AP. 1975. Taphonomy of contemporary and late Cenozoic east African vertebrates.  Ph.D. thesis, University of London, 331 pp.

Hill, AP. 1979.  Disarticulation and scattering of mammal skeletons. Paleontology 5:261-274.

Schafer, W. 1972. Ecology and Palaeoecology of Marine Environments. GY Craig ed. I Oertel translator. The University of Chicago Press

Voorhies, MR. 1969. Taphonomy and population dynamics of an early Pliocene vertebrate fauna, Knox County, Nebraska. University of Wyoming Contributions to Geology, Special Paper1: 1-69.

I’ve been studying an old photo Bob Athen took in 2002 of the bones he and Sonya had collected, spread out in their upstairs hallway and arranged as they originally found them. Like a fortuneteller staring at   the leaves on the bottom of  a teacup, I’m trying to figure out what, if anything, it means about the conditions at the time of death and where we should dig next. 

The large femur (thigh bone) at the bottom of the photo sits furthest downstream, caudal (tail) vertebrae lie upstream, at the top. “Up” and “downstream” are references to today, not necessarily then.  The mix of large and small bones in the photo tells me they haven’t been sorted by flowing water, i.e. the West Tarkio Creek wasn’t flowing over the site in sloth-time.    A platter-sized scapula (shoulder blade) rests on the right and an atlas (the first cervical vertebra or neck bone) on the left–the only bone Bob says he didn’t have to glue back together.  Between them I see many elements from the chest–ribs, sternabrae, sternal ribs (sloths have separate interlocking breast bones instead of a single sternum of cartilage), a collarbone and vertebrae.     It’s no wonder we thought we could recover the whole skeleton in just a couple of weekends. From these bones it looked like the animal was lying in a relatively small area and we only had to dig a little deeper.   Of course, we know now that the main concentration of bones was 8-10 ft. to the north, if you count the adult pelvis as “ground zero,” off the left edge of the photo.    The sliver of bone sitting below and to the right of the scapula is a piece of the toddler’s rib.  We wondered where such a fragile bone could fit in the adult but confirmation of the juvenile had to wait until we finished digging the north side of the creek and discovered the main concentration of juvenile bones about 6-8 ft. to the right (south) of the scapula.  More about that another day.  The bones in the photo must have been moved, but how?  And if we knew how they were moved, could we calculate where the missing bones are?

Water didn’t transport the bones here. Besides the lack of sorting, the bones are sitting in clay with some very fine-grained sand. Any current that could move these bones would have left them buried in sand and gravel.  There was little if any current if this clay settled all around them (Voorhies, 1969).  Obviously the wind didn’t move them either, or gravity. There are only a few possibilities left:  1) they floated here–that is, maybe the sloths died in the water, they bloated and floated, and after the different limbs disarticulated, they were pushed around by the wind and waves and eventually, after the flesh decayed, the bones fell here; 2) predators and/or scavengers dragged the bones here, or; 3) they were accidentally kicked here–underwater or while they sat exposed for a while on the surface of the ground (at the edge of a watering hole?)  Some show more signs of weathering than others.

It’s hard to imagine how three sloths could have died in the water all at the same time, with a flood out of the question (remember, no current).  Could they have fallen through ice and drowned? Starved in a drought?  There are probably clues in the blue-gray clay that surrounds the bones.  Maybe the geochemistry or microfossils hold the answer.  Was this a lake?  The closed oxbow of a river?  We’ve never observed any layers in the clay and there’s no sign that the bones are sitting on a surface that the sloths actually lived on.

Maybe we can define the extent of our deposit with some drilling and coring. If there was a body of water here, how far did it extend? Bob says he’s seen exposures of our blue-gray clay all over the county.  How far would pieces of a sloth float before the bones dropped off?  If the bones floated to these locations, something moved them again.  The packages aren’t anatomically correct–for example, we’re finding ribs, teeth, jaws and arm bones piled together.  Experts would say the bones are “associated but dispersed,” that is they are scattered over an area much larger than an articulated skeleton but can largely be linked to individual animals by anatomical characteristics (Behrensmeyer, 1991).

We’re still wondering about the role of predators here–both as a cause of death and movement.  There are only minor signs of scavenging (unless you count the absence of entire limbs!).  That implies the lack of predator competition at the kill (if it was a kill) and little reason for a wolf, for example, to drag a limb very far, or to gnaw all the missing bones to splinters, but evidence of predators is a topic for another day. Floating away could also account for our missing limbs. In any case, you still need a force like random kicking to explain the skeletal hodgepodge.  But how are you supposed to figure out where a bone could have randomly floated or been kicked?   Maybe we can with the help of some mathematics. . . . Dave    

(to be continued)

References

Behrensmeyer AK. 1991. Terrestrial vertebrate accumulations. In PA Allison and DEG Briggs eds., Taphonomy:  Releasing the Data Locked in the Fossil Record, Ch. 6, Plenum Press, pp. 291-335.

Voorhies, MR. 1969. Taphonomy and population dynamics of an early Pliocene vertebrate fauna, Knox County, Nebraska. University of Wyoming Contributions to Geology, Special Paper 1: 1-69.