In the case of mammals, digestion is commonly defined as the process by which the structures that comprise foods are broken down and transformed into molecules able to be absorbed into the blood stream or, if not, then excreted. Digestion occurs in the digestive tract, which basically consists of a hollow tube and associated solid organs that excrete specialized digestive fluids. The gastrointestinal tube consists of the mouth, esophagus, stomach, small intestine, large intestine (which includes the cecum, colon, and rectum), and anus. The solid elements consist of the liver, pancreas, and gallbladder. Almost all absorption of digestive end-products into the blood stream occurs through the walls of the small intestine, although reabsorption of some water and digestive juices also occurs in the colon.
Different nutrients pose different digestive problems for mammals. Proteins and fats are readily digested, and easily dealt with in the simplist of monogastric digestive tracts typical of most carnivores. Proteins are digested primarily by digestive juices in the stomach and small intestines, including excudates from the pancreas. Fats are digested primarily in the small intestine with the aid of both pancreatic juices and bile acids, the latter processed and concentrated in the gallbladder. Of the carbohydrates, sugar and starch are the easiest to digest. With the addition of key enzymes, even the digestive tract of a monograstric carnivore can deal these relatively simple nutrients (although starch is technically considered to be "complex"). The fructose and glucose that comprise most sugars in fruits are readily absorbed without further digestion, whereas the decomposition of less common sucrose molecules into fructose and glucose requires the admixture of sucrase in the small intestine. Starch digestion requires amylase secreted as part of saliva in the mouth and dextrinase, gulcoamylase, and pancreatic amylase, all of which act to further digest starch in the small intestine.
The digestion of complex structural carbohydrates by mammals is the most difficult of all, and the process that differentiates herbivores from carnivores. Structural carbohydrates are often referred to as fiber, and consist pimarily of cellulose, hemicellulose, and lignin. None of these can be digested unaided by mammals. As a result, most herbivores have entered into a symbiotic relationship with anaerobic bacteria that can digest complex carbohydrates through fermentation. The microbes benefit by obtaining some glucose for food and the residuum of the digesta ends up with the host. All that the host needs to provide is an oxygen-free (anaerobic) environment with lots of fluid, a steady supply of nitrogen and other raw materials, and optimal pH. In almost all heribviores these environments are provided in specialized chambers that are either in front of or after the stomach, that is, either in the foregut or the hindgut. Foregut fermentation almost always occurs in a rumen, whereas hindgut fermentation happens primarily in an enlarged cecum along with some in the colon and small intestine. So...the difference between ruminants (most medium-sized herbivores, including elk, deer, moose, and bison) and non-ruminants (e.g., elephants, hippopotamuses, horses, and rabbits).
Where do bears fit in all of this? They are omnivores, which means they eat a varied diet potentially comprised of either almost wholly meat or almost wholly vegetation. Yet they have the simple monograstric digestive tract of a carnivore that lacks specialized chambers able to sustain anaerobic fermentation by symbiotic microbes (for more on the gut environment of bears, follow this link). More to the point, they lack a rumen and a cecum, although there is evidence of some microbial fermentation in the bears' simple large intestine (see Gut environment). Which means that they obtain little nutritional benefit from the fiber that they ingest, which can comprise 10-30% of the foliage they graze or browse. Which means that their most digestible foods consist of those rich in either digestible protein or fat (realizing that each can come in relatively undigestible forms such as chitin), or containing high concentrations of simpler carboydrates such as fructose, sucrose, or starch.
The graphs at left illustrate the general digestive plight, or strategy, of bears in contrast to fore- and hindgut fermenters (i.e, ruminants and non-ruminants) as well as in comparison to other carnivores. The graph immediately to the left illustrates the relative digestibility of several broad categories of foods by bears and other taxa (with percent digestibility shown as medians and interquartile ranges). For the purposes of this graph, bears are parsed out in different ways, with giant pandas and grizzly bears differentiated for illustrating digestion of foliage and roots, and all bears lumped together as "ursids" for illustrating digestion of meat. These data come from multiple sources.
The basic patterns are pretty obvious. Bears are as well able as any other carnivore to digest most of the meat they eat--around 90% plus. By contrast, grizzly bears digest roughly 20% less of the foliage they consume compared to ruminants, and 10% less when compared to non-ruminant herbivores. Remarkably, giant pandas only digest roughly 20% of the bamboo they ingest, which is only 1/3 of the digestive benefit obtained by ruminants. Recent research suggests that pandas simply pass most of what bamboo they eat through their digestive tract after crushing the stems and foliage to release soluable cell contents. Finally, starchy roots are digested by grizzly bears with about the same efficiency as foliage is digested by ruminants--so a comparatively beneficial vegetal food for bears, at least when reckoned simply in terms of digestibility.
But the other key element of a digestive strategy is not just how well an animal can digest a given gram of ingested food, but also how many grams are being ingested in toto. In other words, an animal can compensate to some extent for low digestibility by increasing the throughput of ingested food. Which seems to be the strategy adopted by bears, especially giant pandas.
The figures at left illustrate this pattern. Again, broad categories of animals are differentiated, with non-ruminant herbivores separated by whether most fermentation of fiber occurs is the cecum versus the colon. Bears (i.e., ursids) are differentiated by whether they are ingesting foliage and fruit versus wholly meat. The top figure shows the rate at which these different categories of animals ingest food, standardized to metabolism-corrected body mass, whereas the bottom graph shows the mean time that digesta is retained in the digestive tract (i.e., gut; the inverse of the rapidity of transit).
From the figures above, it appears that, when possible, bears ingest vegetal material at a higher rate and retain it for a far shorter period of time compared to specialized herbivores, especially in contrast to foregut fermenters (i.e., ruminants). This would partly compensate for the lower efficiency with which bears digest most vegetal food. By contrast, bears ingest meat at a slow rate and retain it for roughly twice as long as they do their vegetal food. Which makes sense. For the high digestibility of meat to be realized, bears probably need to retain it longer in the digestive tract, but still not as long as herbivores dependent on fermentation retain foliage or browse. Meat also probably passes through the gut more slowly simply because there is less accompanying fiber to hasten it along compared to when bears eat vegetation.
The implications of these patterns for bears and bear foraging seem pretty obvious. Bears should prefer meat whenever they can get it, at least until sated, and up until a need to balance nutrient intake comes into play (see Protein & energy effects). Beyond that, roots (and berries) should be preferred vegetal foods, but only if the energy required for acquistion does not unfavorably alter the overall energetic equation--which, in the case of roots, is probably often the case because of the potentially considerable costs of excavation. Finally, bears should be able to profit from grazing only when they have access to large amounts of readily acquired and comparatively digestible foliage. And, as shown below, digestibilities of foliage can vary widely, not only among sites, but also among plant species and seasons.
Comparative digestion in bears & herbivores
In keeping with the broad patterns described above, the digestibilities of specific bear foods vary widely. Emblematic of this, the graph immediately to the right shows the percent of energy contained in different foods that is digestable by grizzly bears (the black, gray, and white dots). The varying shades of gray, from black to white, correspond to digestibilities during different seasons in instances where there is documented seasonal variability: black for spring, dark gray for estrus, light gray for early hyperphagia, and white for late hyperphagia. The reddish dots represent the percent of each food that is comprised of protein, again with seasonal variation denoted by varying shades: bright red for year-round or spring values; burgundy for mid-season; and white for late-season. All of these foods are specific to the Yellowstone ecosystem.
So...consistent with what I presented at the top of this page, meat from any source is more digestible than other types of food. Roots, insects, and fruits and seeds are comparable in terms of digestability, but with roots and fruits offering far less protein. Most of the digestable energy in these vegetal foods is contained in sugars and starches, with the proviso that much of the protein in ants, in particular, is bound up in chitin. Finally, the digestible energy in foliage varies widely, with forbs such as clover, fireweed, and dandelion offering the most, and elk thistle, horsetail, and grasses and sedges (i.e., graminoids) the least.
Having noted the differences in digestible energy offered by many forb species in contrast to grasses, an obvious question arises. Why do grizzlies consume so few species of forbs? Forbs, on average, have a higher protein and lower fiber content compared to graminoids, and consumption of graminioids by bears seems to be dictated more by plant architecture and fiber content, as such, rather than by the identity of any given species. One might expect bears to focus exclusively and indiscriminately on grazing forbs rather than graminoids.
Based on a number of observations, I suspect that the highly varied and selective consumption of forbs by grizzlies is a reflection of variation in secondary chemical compounds. Whereas graminoids defend themselves from grazing primarily by producing high concentrations of fiber and silica, forbs defend themselves primarily through the production of various toxic, mildly toxic, or digestion-impairing chemicals. Tannins are a classic example in the mature foliage of forbs such as fireweed. On the other hand, some compounds produced by forbs are probably beneficial, even therapeutic. Clover and horsetail contain several compounds with potentially beneficial effects, and others with the potential to harm. For example, thiaminase in horsetail can cause thiamine deficiencies in animals that heavily graze it, resulting in edema and lack of motor control. Given all of this, I suspect that grizzlies only consume those few forb species that contain a balance of secondary compounds posing them no net harm, or perhaps even offering them medicinal benefits.
As a final note on the figure above: The pinkish horizontal band corresponds to the optimal level of protein in bear diets (see Protein and energy effects). The point being that grizzlies would be hard-pressed to maintain an optimal level of protein intake if they subsisted solely on vegetal foods, especially roots, fruits, and seeds.
Given the considerable variation in digestibility of different foods, one might expect that the residues of foods in excreted feces would be a poor reflection of what bears ate. Meat would be considerably under-represented, the contribution of foliage grossly inflated, and berries and roots somewhere in between. And, in fact, this is the case.
The height of the bars in the graph to the left is proportional to the factors needed to correct the fractions of each food found in feces to more closely approximate the volume that was ingested. For example, most foliage would be multiplied by a factor less than 0.5; some roots, berries, and seeds would be multiplied by a factor somewhere between 1 and 2; and animals by factors ranging in value from roughly 1.5 to over 12.
Scat correction factors
One obvious point to made here is that scat correction factors are more or less correlated with dry matter digestibilities. But far from perfectly. Differences arise from the fact that correction factors reflect not only digestibilities but also the many factors that affect the shear physical bulk of different foods in scats. Bulk can be affected by the extent to which foods are masticated and structural features destroyed. In the case of animal foods, bulk is affected in a major way by the amount of ingested hair, hide, and bone, as reflected in the range of correction factors that apply to mule deer tissue, from roughly 1.5 to over 3. And, certainly, the off-the-charts correction factor applicable to ground squirrels has little to do with any greater digestibility of ground squirrel tissues compared to meat from other sources.
There are implications of all this for anyone who wants to estimate the contributions of different foods to total digested energy, working from scat contents backwards. Logically, you would first apply the correction factors to come up with an estimate of ingested dry matter. Then you would apply percent digestibilities to these estimates of ingested dry matter to come up with an estimate of contributions to total dietary energy. Having laid out this formula, there are more complications to the calculations that ideally would to be considered. But such would be the bare bones.
For more on the gut environment of bears, click this link: