We evolutionary biologists have one advantage over other biologists: we’re constantly encountering bizarre features of nature showing the almost unbelievable variety of adaptations that can be produced by natural selection. This means that we’re constantly getting little frissons of pleasure, and I get at least one of these a month. The most recent came from a new paper in New Phytologist by Chin et al. These authors make a pretty compelling case that the morphology of three species of pitcher plants in Borneo has evolved to not only catch the droppings of tree shews (and use them as nutrients), but to compel the shrews to defecate (and probably urinate) in their pitchers. In other words, they’ve evolved to be shrew toilets.
Working on Mount Kinabalu in Borneo, Lee et al. discovered that three species of pitcher plants in the genus Nepenthes (N. lowii, N. macrophylla, and N. rajah, the world’s largest carnivorous plant, shown in Fig. 2), have evolved features that make them attractive to treeshrews of the genus Tupaia. (Treeshrews are neither rodents nor shrews; they’re a group more closely related to primates than to rodents). The plants have recurved “lids” that produce a sweet substance that the tree shews lap up while sitting astride the pitchers.
The authors found that, depending on the species of plant, between 60% and 90% of the pitchers contained fecal pellets from treeshrews. Video cameras placed near pitchers of N. rajah captured 7 treeshrew visits, lasting an average of 24 seconds, and one of these showed the beast crapping into the pitcher. Previous analysis (Clarke et al. 2009) showed that these pellets provide a large fraction (58-90%) of the nitrogen needed by N lowii.
Geometric analysis of the three feces-trapping species showed that the leaf morphology has evolved in such a way that forces the shrews to sit astride the plant while eating its exudate (Fig. 2). The three species in which feces were observed all have a combination of lid angle, lid concavity, and pitcher width that forces the shrews (given their body size, which was also measured) to straddle the plant if it wants access to the entire lid surface. Other species, and pitchers of immature plants in the three “target” species, don’t have this combination of traits, and although treeshrews will sometimes lick their lids, they don’t defecate in them.

Figure 1 (from Chin et al.) The senior author measuring a pitcher of N. macrophylla. Bar is 5 cm (about 2 inches).

Figure 2 (from Chin et al.). Images from videocam recordings of N. rajah pitchers. Left: T. montana stradding a pitcher and eating its secretions. Note the animal’s butt inside the pitcher. Right: treeshrew feces in a pitcher. Scale: 5 cm.
Now this coincidence between treeshrew size and plant geometry may be a coincidence rather than an adaptation, but it’s a good guess that, given the frequency of fecal-pellet deposition in the pitchers and the important contribution they make to the plant’s acquisition of nitrogen, as well as the fact that the three species of pellet-using pitchers are evolutionary outliers among Nepenthes in their large size and lid configuration, the pitchers have evolved to be shrew loos. In the earlier paper by Clarke et al., the authors posit that, because some treeshrews scent mark the plants by urinating on them, the pitchers could also be using this urine.
The authors describe the whole situation as “an extraordinary example of co-evolution and specialization” but I reserve the the term “co-evolution” for cases, like figs and their fig-wasp pollinators, in which both species have evolved because of their interactions. In this case there’s no evidence that treeshrews have themselves undergone any evolutionary change since they started using pitcher plants, although this is possible. The shrews, for example, could have a genetically based propensity to seek out pitcher plants. But we don’t know if this is the case.
Fig. 3. I can haz privacy? A mountain treeshrew doing its business on a pitcher. Photograph by Ch’ien C. Lee.
Pitcher plants are a remarkable example of convergent evolution: the evolutionary modification of leaves into pitchers that trap insects or other nutrient-providing stuff has occurred three times in three independent groups, the Sarraceniaceae (North and South America), the Nepenthaceae (tropical Asia), and the Cephalotaceae (Australia).
And, finally, there’s one species of ant, Campanotus schmitzi, that dives into the pitcher-plant liquid and retrieves drowned insects as well as living mosquito larvae. These ants can actually swim, remaining submerged for up to 30 seconds! The authors note that it can then take an ant up to 12 hours to haul its prey out of the fluid and up to the pitcher mouth.
Clarke and Kitching (1995), the authors of the study, suggest that this is a mutualism between ant and plant since the ants prevent too many arthropods from accumulating and putrifying inside the pitcher—something that disrupts the plant’s ability to digest them.
h/t: Matthew “Swammerdam” Cobb
________
Chin, L., J. A. Moran, and C. Clarke. 2010. Trap geometry in the giant montane pitcher plant species from Borneo is a function of tree shrew body size. New Phytologist, in press (early view).
Clarke, C. M. and R. L. Kitching. 1995. Swimming ants and pitcher plants: a unique ant-plant interaction from Borneo. J. Tropical Ecology 11:589-602.
Clarke, C. M., U. Bauer, C. C. Lee, A. A. Tuen, K. Rembold, and J. A. Moran. 2009. Tree shrew lavatories: a novel nitrogen sequestration strategy in a tropical pitcher plant. Biology Letters 5:632-635.





