Front limbs


Of all the morphologic features that typify bears, the front limbs and associated skeletal infrastructure are the most distinctive. They are also diagnostic of the bear life strategy (see Life strategy). No other terrestrial vertebrate of its size--certainly no other large carnivore--has front limbs that are as flexible, powerfully built, and mounted with such dexterous paws. Nor do any comparable-sized carnivores have such out-sized claws...claws which are clearly "designed" to be powered by the muscular arms and shoulders to either climb trees, extract food from a durable matrix (i.e., dig), or grapple with and subdue large prey such as seals, moose, and elk. What follows is a summary of the evidence produced over the years elaborating on and substantiating the preceding thumbnail sketch. You will have to forgive me for the abundance that follows, but it is reflective of the extent to which I see this aspect of bear morphology as key to understanding the overall bear life strategy--as well as niche. 

Ihe results at left come from a seminal paper published by Alberto Martin-Serra and his colleagues in 2014. His analyses delve into patterns of variation in post-cranial skeletal structure among carnivores. Some of the results are relatively arcane, as is often the case in morphology. But the overall results insofar as bears are concerned are pretty straight-forward.


Each of the graphs at left represents major coordinated trends in the shape of two important front-limb bones--the radius at top, and the scapula (or shoulder blade) in the bottom two graphs.  This variation is after accounting for the effects of body size, as such. Each dot represents an individual specimen, with those of bears denoted by brown dots and those of all other carnivore species by gray dots. The bones illustrated along each axis portray the shape that is associated with extreme values along each gradient. "PC I" invariably represents the dominant theme of variation, "PC II" the secondary theme, and "PC III" the tertiary theme. The typifying characteristics of the quadrant occupied by the bear specimens are described in the corresponding corner.


The key results here are: first, that bears are clearly differentiated from all other carnivores when it comes to shape of the radius and scapula, especially so for the scapula; and, second, that the main differentiating theme is captured by the term "robust." In other words, even after controlling for the effects of body size, bears have stout and strongly built front limb bones, most remarkably so in the case of the scapula. Which implies that these robust bones are built to support disproportionately strong front-limb muscles. Which further implies that the "bear niche" is typified by activities that require deployment of strong front legs anchored to a strong scapula. 

The graphic above (thanks to a 1949 paper by Dwight Davis) emphasizes the concluding point made at left, more specifically that the stout skeletal infrastructure of the front limbs in bears--most especially the scapula--is host to proportionately robust muscles. Each color in the bars above corresponds to the proportion of all muscle mass in the shoulder and arms of lions (Panthera), wolves (Canis), and bears (Ursus) consisting of the muscle group identified to the right of the three bars. The key point here is that scapular muscles (anchored to the scapula) account for a disproportional amount of muscle mass in bears. This is especially true for the brown or grizzly bear, which accounts, in part, for the "hump" that typifies this species.


The figures above illustrate the flexibility (or dexterousness) of the forelimb in bears. The bar-shaped summary values at upper left derive from the work of Andy Iwaniuk, who devised a scheme for scoring dexterity and then applied it to a number of carnivore species through direct observations of zoo animals. The result of his work is pretty obvious. Bears (i.e., Ursids) are by far the most dextrrous of large carnivores, whether reckoned in terms of "distal" or "total" forepaw dexterity--comparable to the dexterity of the much smaller procyonids (e.g., raccoons). This is remarkable given that the bears have to support much more weight on their front legs and paws than do raccoons and their relatives.


The results above right relate pretty directly to the idea of "dexterity." Here, based on the work of Ki Andersson, main themes in variation of the shape of the ulna (upper arm bone) where it intersects the elbow joint are plotted against body mass, again restricted to carnivores. Each dot represents a species, with bears denoted by brown dots, canids (dogs) by dark red dots, and felids (or cats) by the orange dots. "GB" denotes "grizzly bears" (Ursus arctos). The key distinction noted in this graph is between the "cursors" and "grapplers"--the cursors being those that are built to run long distances efficiently and the grapplers those built to grab and subdue large prey after a short rush. Or, perhaps, simply dig roots. Bears are clearly among the grapplers, which is wholly consistent with a dexterous forelimb and paw.

Size and shape of the olecranon process have been singled out by morphologists as being especially indicative of how animals move and, related to that, the basic configuration of their posture. The olecranon process is at the end of the ulna, where it is an integral part of the elbow as well as the anchor for the triceps muscles (the ulna is paired with the radius in the lower arm). The relationship of the olecranon process to the "semilunar notch" and to the main axis of the ulnar bone are particularly signifying. The semilunar notch is the half-sphere hollow in the side of the ulna that hosts the end of the humerus. This is where the radius and ulna pivot around the end of the humerus, which is the upper arm bone.


A lesser capacity for sustained speed and a greater capacity for bearing the load of a heavy body are indicated by: (1) a relatively short olecranon process; and (2) a wide angle between the main axis of the ulna below the notch and the main axis of the olecranon process--which indicates a rearward facing process. The basic difference is characterized as either "cursorial" (built for sustained speed) or "ambulatory" (built to ambulate).


The figure above again draws on the work of Martin-Serra, but focused here on the shape of the ulna. This shape-related variation is what remains after accounting for the effects of body size. As you can see, the bears (dark brown dots) are differentiated from all other carnivores primarily by the shape of the olecarnon process (along PC II). More specifically, bears have relatively the shortest and most rearward (or caudal)-facing process, which signifies a less important role for the triceps (versus the biceps)...which signifies, in turn, a design to hold up under various load stresses rather than to efficiently move about over long distances.


The figures above, left, show how the various bear species (each denoted by an orange dot) plot relative to other carnivores when comparing body mass to proportional length (top) and orientation (bottom) of the olecranon process. These data come mostly from Blaire van Valkenbugh, but also from other researchers. Orientation is measured in terms of degrees of inclination toward the tail-end (caudal end) of the body. These results contrast with the results shown in the figure to the immediately above, which depict variation left over after controlling for the dominant effect of body size.


Considering all of the carnivores, the basic theme is one of an increasing backward angle (but at a decreasing rate) and increasing relative length (at an increasing rate) as body mass increases. Generally speaking, the bears cluster with other carnivores of the same size. They tend to have a similar configuration of the olecranon process, roughly of the sort you would expect given how big they are.


The one exception, though, is the brown or grizzly bear (Urar, Ursus arctos). The grizzly bear has a lesser backward orientation to the olecranon relative to what you would expect by its body size, and the longest olecranon process relative to the ulnar shaft length of any carnivore. This last feature is noteworthy given that long olecrana have been associated with animals who spend a lot of time digging--which is consistent with the notion that adaptations for digging are a key feature of the grizzly bear niche. Which is consistent with the robust scapula and well-developed scapular muscles of most bears (see above), which also seem designed for activities such as digging.


The results from Martin-Serra's work (above right) could be read as suggesting that all bears have a short olecranon--in contrast to the result just described above for grizzly bears. Reconciliation of this apparent discrepancy could involve the considerable statistical gymnastics that Martin-Serra went through to "control for" effects of body size and evolutionary relatedness. It could also have something to do with the brown dot in his graph denoting the bear species with the lowest score on his PC II axis. Although the species associated with this dot was not identified, it may represent brown/grizzly bears and, if so, be an outlier for this family, at least in the context of Martin-Serra's work.

Olecranon process of the ulna

Running speed

The comparatively slow running speed of bears lends additional support to the theme that bears have heavily built forelimbs and highly flexible paws that are "designed" to facilitate activities such as climbing or digging rather than moving long (or even short) distances efficiently and rapidly.


The slow speed of bears is illustrated graphically immediately above. Maximum running speed of numerous terrestrial mammals is plotted relative to body features that have previously been correlated with speed: total mass (or overall size) and length of the fore- and hind-limbs. The general relationship is one of increasing absolute speed with size, and then a decline. Increasing limb length would likely produce a longer stride--and, hence, speed--but only up to a point. That point being where increasing bulk would (1) mandate that bones accommodate load stresses rather than deliver speed, as well as (2) slow the gait because of the inertia of increasingly massive limbs.


But notice the bears (brown dots). They fall well below the trend line in all three relationships, suggesting that at least the three bear species represented here are slow given their size (roughly 50 kph [30 mph], maximum speed).  This is not surprising in light of the information summarized above, all of which paints a picture of limbs that provide ample strength and flexibility for climbing, digging, and grappling, rather than ample stored and released energy for rapid and efficient movement. If you watch a bear run (albeit faster than most people), you will probably be struck by how inefficient the whole process seems, especially compared to a deer or an elk. 


The final topic that I cover here under "forelimbs" is claws (or the ungal: highly-modified terminal phalanges). As a general point, bear claws are not retractable like those of cats--they are more like the non-retractable claws of dogs and the larger members of the weasel family. Looking at the graph immediately to the upper left (again, thanks to Blaire van Valkenburgh), bears fall out in a distinctive quadrant when you plot the characteristic depth of a claw against its curvature--which signifies how stout versus elongate it is. Considering all of the carnivores, there is a general trend from deep slightly curved claws, characteristic of the cats (the orange dots), to elongate claws (curved, but less deep) more common of canids (dark red dots) and bears (brown dots). The inset profiles of various claw shapes illustrate the extremes.


So...notice, again, the brown or grizzly bear (Uar, U. arctos). Next to the badger (the gray dot farther to the right and below), grizzlies have the most elongate claws of all the carnivores shown here. The graph identifies the realm of "scratch diggers"--animals that live aboveground but dig for a significant part of their food, whether rodents, squirrels, or roots. Grizzly bears fall within this type. Added to the evidence presented above, this bit of data pretty conclusively establishes that Ursus arctos is well-built to excavate roots and rodents--that, in fact, the ability to excavate such foods probably defined a significant part of the evolutionary imperative that led to the emergence of this species (more on this under Foods and History). 


Parenthetically, relative claw length seems to systematically vary among brown bear genetic lineages (clades; see Evolution). The map above shows the length of claws relative to the length of skulls for various populations of brown bears in Eurasia. The claw profiles are scaled to this value, which is presented in numeric form immediately above. This information is overlain on a map of brown bear clades and skull sizes (see Body size) for reference. The data come from publications by Bjorn Kurten and Sergei Ognev that date to the middle part of the last century--when biologists were interested in compiling such arcana. Brown bears in Europe, the Caucasus, and among the salmon-eating bears along the Pacific coast had the smallest claws. Everywhere else the claws were relatively larger.


But the key pattern in this map relates to the exceptionally large claws of the bears living on and near the Tibetan Plateau, part of the relatively limited Clades 5 and 6. These clades were apparently isolated at high elevations during the late Pleistocene, which may explain why they are so distinct in appearance--including claw size. But there is also a likely link to diet. Brown bears on the Tibetan Plateau rely to an exceptional degree on excavated foods, including the Plateau pika, the Tibetan marmot, as well as various roots (see Foods). On the other hand, brown bears living in Europe and among salmon along the Pacific coast don't dig much at all--other than into fairly soft sand (the coast) or ant hills (Europe). So...the importance of claws for digging is potentially evident even in variation of these appendages among brown bear populations in Eurasia.