Introduction by Jerry: It’s been known for a long time that three-toed sloths (genus Bradypus, several species) have the odd habit of descending about once a week from the trees in which they live, defecating on the ground at the base of the trees, burying their feces under leaves, and then climbing back up the trees, whose leaves are their food.  I’ve posted about this behavior twice before, once showing an Attenborough video, with the Great Man noting that the behavior is a mystery, and the other with a note by Manchester researcher Becky Cliffe, who suggested that the behavior is involved in mating, perhaps allowing other sloths to find females. (There are at least two other hypotheses as well.)

Curiously, this behavior is not known in two-toed sloths (genus Choloepus), which defecate from high up in the trees, giving away their location when the feces hit the ground (it’s been suggested that this behavior alerts predators).  One interesting fact mentioned in a new paper on this behavior by Jonathan Pauli et al. (reference and link below; download may be free, but if not, some judicious inquiry will yield you a pdf) is that two- and three-toed sloths are very distant relatives, separated by 40 million years of evolution. That’s almost as long as dogs and bears have been separated!

At any rate, reader Bruce Lyon perused this paper and emailed me his summary of it, noting its hypothesis about why three-toed sloths undergo such a long and arduous slog to have a poo. (It’s also a dangerous slog, because, as Bruce notes below, a large portion of sloth mortality occurs when they’re pooping.)  We still don’t know the reasons for this bizarre behavior, which is obviously evolved, but Bruce summarizes an interesting theory supported by some data in the paper. He also supplied several pictures of sloths, both two- and three-toed, that he took in Costa Rica.

After Bruce’s short essay I’ll add a few remarks of my own about the evolution of this trait.

Of sloths and moths

by Bruce Lyon

I spend two weeks each winter with my family in Northwestern Costa Rica. We often see two species of sloths—Hoffman’s Two-toed Sloth [Choloepus hoffmanni] and the Three–toed Sloth. As the names suggest, they differ in number of toes, which can be clearly seen in some of the photos below (count the long claws). They also differ in general appearance: the two-toed sloth has lighter fur and a pig-like nose, while the three-toed has a bandit-like face mask and has a perpetual smile. Some of the animals we see are fairly strongly tinged with a green color, caused by algae growing in their fur. [JAC: it’s been suggested this green color helps camouflage the animal, hiding it from predators.]

Not knowing much about sloths, I had always assumed they were just unkempt. Not so. An interesting paper just out in Proceedings of the Royal Society by Jonathan Pauli and colleagues set me straight on the green fur business, and also has some interesting ideas about the sloth’s bizarre habit of descending from trees to defecate.

The paper combines interesting facts, new observations, and speculation. Sloths are famous for being incredibly slow–paced animals, giving rise to the expression ‘sloth’ that denotes laziness. Sloths, however, are not lazy but slow—the three-toed sloth has a very low metabolic rate and the slowest rate of digestion of any mammal, possibly connected to their specialized diet of nutritionally poor and toxic leaves. [JAC note: Becky’s post says that it take about 31 days from ingestion of leaves to excretion.]

Three-toed sloths also have the curious habit of descending from the canopy weekly to defecate—the animal digs a shallow hole, goes about its business, covers the dung and then climbs back up into the tree. Given the low metabolic rate, spending all the effort to climb out the tree turns out to be expensive (estimated at 8% of the daily energy budget) and it is also risky (half of all adult sloth mortality in one study took place near or on the ground).

Both species harbor a diverse collection of organisms, including green algae (hence the greenish tinge on the fur) that appear to be facilitated by the structure of the sloth’s hair that retains moisture. There are also pyralid moths that live in the fur, and these can only reproduce in the sloth dung. The new study hypothesizes that the curious defecation habits of the three-toed sloth may reflect a mutualism between the sloths and the moths.

The authors collected sloths of both species and showed that for the three-toed (but not two-toed) sloth, the number of moths infesting each animal was correlated with the concentration of nitrogen on the fur, which was correlated with chlorophyll levels, an indication of the concentration of algae. The paper also provides some evidence that the sloths are eating the algae, which is rich in carbohydrates. The speculative part is that by climbing down the ground to defecate, a sloth enhances the success of the moth (this may be known from other studies) and most importantly, that the increased number of moths that emerge from the dung go back and specifically benefit the same sloth. The paper has gotten lots of press and a fuller account can be found over at Ed Yong’s website. [JAC: Bruce doesn’t mention that the mutualism is good for the sloth because the moths that hatch from sloth poop carry both organic detritus and fungi back to the sloths where they are obligate residents—moths mate only on sloths, and are not removed by grooming—and that those fungi provide nutrients that promote the growth of algae.]

[Bruce’s photos; his comments indented]

Mother and baby Hoffman’s Two-toed Sloth:

Another mother and baby two-toed perched at eye level! Note the greenish tinge to the fur.

[JAC: The paper suggests that babies get the algae from their mother. Three-toed sloths, unlike two-toed ones, also have special cracks in their fur that collect rainwater and promote the growth of algae. This is likely to be an adaptation: a morphological feature of the animal that helps turn it into an algae garden!]

Three-toed sloth in a Cecropia tree. My impression is that these sloths love this tree species, a weedy fast growing species that favors open areas. Check out the interesting black and orange dorsal stripe. This stripe is found only in males, and is therefore probably the result of sexual selection.

Jerry’s addendum:

Here are some data from the paper showing that the three-toed sloths are moth-ier and have more nutrients in their fur than do their distant two-toed relatives. They do ingest these nutrients when grooming. This is Figure 1 with its caption, as well as a photo of the pyramid moth (Cryptoses spp.) that is an obligate resident (probably a mutualist) on the three-toed sloth.

Figure 1. Both (a) three- and (b) two-toed sloths harbour a diverse ecosystem in their fur. (c–e) The more sedentary three-toed sloths (black bars) possessed (c) a greater number of moths, (d ) more inorganic nitrogen in the form of NH4+ and (e) greater algal biomass on their fur compared with two-toed sloths (grey bars). Error bars represent +1 s.e.; *p , 0.05, **p , 0.001.

Here’s a generalized figure from the paper that summarizes the existing mutualism between moth and three-toed sloth:

Figure 3. Postulated linked mutualisms (+) among sloths, moths and algae: (a) sloths descend their tree to defecate, and deliver gravid female sloth moths (+) to oviposition sites in their dung; (b) larval moths are copraphagous and as adults seek sloths in the canopy; (c) moths represent portals for nutrients, and via decomposition and mineralization by detritivores increase inorganic nitrogen levels in sloth fur, which fuels algal (+) growth, and (d ) sloths (+) then consume these algae-gardens, presumably to augment their limited diet.

I have a few caveats about this paper, but first my favorite sentence from the methods and materials section of the paper and then a note about sloth-keeping.

“We cut a lock of hair from the dorsum of each sloth and collected all moths from the sloth with an invertebrate vacuum.”

For some reason that sentence makes me laugh: it conjures up the image of trimming a sloth’s back, vacuuming it, and the suggestion that vacuums don’t have backbones.

One interesting suggestion from the authors is that the difficulty of rearing three-toed sloths in captivity (as opposed to two-toed sloths) is because the keepers remove the feces, breaking the cycle of mutualism with the moth that helps nourish the sloth.

Finally, here are two reasons why I find the “moth mutualism” hypothesis either problematic or far from “proven”.

1.  The mortality of three-toed sloths while defecating is substantial: as Bruce notes, half of all sloth mortality occurs on or near the ground. That’s a substantial evolutionary cost, and if the behavior of defecating on the ground is adaptive, it has to provide advantages that more than outweigh this cost. It’s hard to imagine that this mutualism, which appears to provide the sloth with both nutrients and camouflage, more than makes up for the halving of fitness associated with ground defecation. The authors certainly haven’t shown this: they’ve shown that there appears to be a nutritional advantage to harboring moths. But they haven’t shown that this advantages also enhances reproduction, much less near two-fold, nor have they shown anything about camouflage.

On the other hand, if the defecation somehow enhances mating, by alerting sloths to the presence of an opposite-sex sloth nearby, it’s easier to see a big advantage. Becky’s post noted this:

Female sloths have a very regular 28 day reproductive cycle, during which, they are in heat for 10 days. During these 10 days,they actually let out regular high pitched screams that attract males from up to 700m away, and they descend to the same spot at the base of the same tree daily to go poo/pee (even if it is just the tiniest amount!)

Finding a mate, of course, is a behavior that immensely enhances fitness; if you don’t find one, you’re a genetic loser. To me, this seems to confer a stronger evolutionary advantage to ground-pooing than does the mutualism with the moths (of course, both advantages could work together).

2. If mutualisms with moths was the main evolutionary impetus for ground-pooing, how did its evolution begin? This requires that the first individual that developed the habit of descending from the tree to defecate on the ground would gain an advantage over his (or her) fellow sloths. But that would occur only if the moths that hatched form the feces (and remember, at this stage the moths were not obligate residents on sloths), went back to the very same sloth from whose poop they hatched and helped the algae colonize its fur. (Bruce notes this potential problem above.) If they went to other sloths, or their original breeding/mating habitats, then there would be no specific advantage to the first sloth who put itself in danger by pooping on the ground. (I’m assuming that pooping from the treetops is the ancestral condition). In other words the other sloths who defecated from the trees would be “free riders,” gaining the advantage of moth colonization without the dangers.

The author’s hypothesis, then, implicitly presumes that sloth density is so low that any moths that hatch from sloth dung would perforce go to the same sloth who produced the dung. Well, that may be possible, for I don’t think sloth density is very high in the neotropics.

Nevertheless, I would suggest that the stronger evolutionary “force” promoting this behavior is not crypsis/nutrition but reproduction, as Becky suggested.  That makes more sense to me for two reasons. Finding mates seems to confer a stronger selective advantage than does a marginal greening of the fur, and because at the beginning of the trait’s evolution, moths were not limited to sloths and could breed anywhere, both on other sloths and at other locales where they originally bred.

Perhaps the mutualism with moths nudged the ground-pooping trait along, but I suggest that it’s a “spandrel.” That is, after sloths had evolved to defecate on the ground for reproductive reasons, the moths took advantage of this egg-laying site and secondarily colonized the sloths’ fur. That was good for the moths, for it would put them close to a source of food for their larvae when the sloth produced its dung. And the benefits to the sloth would be incidental, but perhaps of a size that would strengthen the selection for ground-pooping and lead to morphological adaptations like the cracks in the fur.

At any rate, the authors’ suggestion, which is supported by some data, is interesting; but the evolution of ground-pooping remains a mystery. A behavior with such a strong reproductive cost much have some compensating advantage to explain its evolution, and we don’t yet know what that was. And, no matter what Larry Moran says, I don’t think this behavior is a result of genetic drift, for it’s complex, involving climbing down the trunk once a week and then burying the feces.