Ants mediate nitrogen relations of an epiphytic fern

Henry David Thoreau once said that ‘nature made ferns for pure leaves to show what she could do in that line’ (Myerson, 1992). Indeed, ferns have thoroughly explored the diversity of leaf function and use the laminar surface as a site for both carbon fixation and reproduction. The fern’s lack of flowers has led most to overlook the potential for fern–animal interactions. Yet recent discoveries of lepidopteron soral crypsis in several tropical ferns (Barker et al., 2005) combined with an increased understanding of fern–herbivore interactions (Balick et al., 1978; Auerbach & Hendrix, 1980; Weintraub et al., 1995; Jensen & Holman, 2000; Mehltreter et al., 2003), and the presence of myrmecotrophy in some species (Rashbrook et al., 1992), demonstrate that fern–animal interactions may be more common than once thought. Myrmecotrophy is an intriguing and important plant–animal relationship that has significant consequences for plant nutrition, protection, and ecosystem-level processes (Solano & Dejean, 2004; Fiedler et al., 2007; Palmer & Brody, 2007; Sternberg et al., 2007). In the myrmecophytic relationship, host plants typically provide food resources, for example elaiosomes, extrafloral nectaries (Beattie, 1989), and/or suitable nesting spaces (myrmecodomatia) for the ant visitor. In return, it is assumed that ants protect their host plants by removing herbivores and pathogens, attacking competing vegetation (Janzen, 1969) and supplying nutrients (Kaufmann & Maschwitz, 2006). Myrmecotrophy has been reported in ferns, and species of Solanopteris (Forel, 1904; Gómez, 1974, 1977), Lecanopteris (Gay, 1991, 1993b; Gay & Hensen, 1992), and Polypodium (Koptur et al., 1998) are known to produce potato-like tubers that function as domatia. Limited evidence suggests that nesting ants act to protect host ferns as in the case of Solanopteris brunei, where Azteca ants become quite aggressive when their host plant is disturbed (Gómez, 1974; and personal observation). Perhaps one of the best known temperate fern–ant relationships occurs in the widespread bracken fern (Pteridium aquilinum). This species produces foliar nectaries, and several studies have examined the ecology of this phenomenon, finding limited to no influence of ants on the host plant and vice versa (Tempel, 1983; Heads & Lawton, 1984; Lawton & Heads, 1984; Heads, 1986; Rashbrook et al., 1992). Such results have added to the general rejection of the importance of fern–ant relationships. Apart from serving as a protective mechanism, ants may also contribute to host plant nutrition. While there are a large number of papers dealing with ant gardens and host plant interactions (Kaufmann & Maschwitz, 2006), quantification of nutrient exchange between ants and their plant host in natural conditions has not been widely demonstrated in epiphytic taxa and less so in ferns. Gay (1993a) conducted an elegant series of labeled nitrogen (N) laboratory experiments clearly demonstrating nutrient exchange between ants and host plants in the fern genus Lecanopteris. In this study, ant-derived nutrients were taken up through the inner walls of the domatia and, in at least one species, via roots produced inside such domatia. While Gay (1993a) demonstrated uptake, the study did not demonstrate the relative importance of this relationship to the overall nutrient budget of the host plant. In another example of N exchange in a myrmecophytic epiphyte, Treseder et al. (1995) demonstrated that the Asclepiad Dischidia major (Vahl) Merr. may derive up to 39% of its carbon and 29% of its N budget from ants that it hosts in specialized domatia (Treseder et al., 1995). Here we describe a previously unknown cryptic relationship between the fern Antrophyum lanceolatum and the ant Pheidole flavens and comment on its ecological significance.