On the uniqueness of fern–insect interactions

Abstract

Ferns are believed to have first emerged at least 423 million years ago (Ma) (Nitta et al., 2022). By contrast, the majority of flowering plant diversity appeared 50–100 Ma (Benton et al., 2022). This much earlier appearance of ferns allows for the possibility of a long and unique evolutionary history, first running before, and then parallel with that of angiosperm–herbivorous insect interactions. There was a burst of fern diversification contemporaneous with that of flowering plants that led to 80% of the current fern diversity. The earliest fern and fern relatives were subject to herbivory only by insects with sucking and piercing mouthparts, but by the end of the Pennsylvanian, some 300 Ma, most major types of herbivore feeding had evolved to include chewing, sucking and piercing, boring, galling, and spore consumption (Labandeira, 2002).

Initial analyses in the 1970–1980's suggested that present-day ferns, relative to angiosperms, are underutilized by herbivorous insects (Cooper-Driver, 1978; Auerbach & Hendrix, 1980; Hendrix, 1980). This underutilization was both surprising and perplexing in light of a priori predictions based on available information (Hendrix, 1980). First, because ferns are relatively ancient, we would expect them to be attacked by an insect fauna more diverse than those found on angiosperms, having accumulated more species over their longer evolutionary history (the ‘time hypothesis’ of Southwood) (Hendrix, 1980). Second, we would expect that fern insect faunas would be comprised of more ancient insect taxa of herbivorous insects than angiosperms (Cooper-Driver, 1978). Finally, initial studies revealed a lower diversity of defensive compounds in ferns (Cooper-Driver, 1978), again leading to the prediction of a more diverse herbivore fauna on ferns and greater damage compared to angiosperms. Contrary to expectations, studies found equal or fewer numbers of insect species attacking ferns than angiosperms of similar architecture (Auerbach & Hendrix, 1980). Furthermore, fern herbivore faunas did not appear to include ancient insect taxa (Hendrix, 1980; but see Tahvanainen & Niemelä, 1987), and damage levels were similar in ferns and angiosperms (Hendrix & Marquis, 1983).

Porto et al. address many of these issues and more. They found that different orders of herbivorous insects cluster within the fern phylogeny, with Diptera (flies) being the most specialized. Along with Diptera, Lepidoptera (butterflies and moths), Coleoptera (beetles) and fern-associated Hymenoptera (sawflies and ants) are more specialized, while Hemiptera (aphids and aphid relatives) and Orthoptera (grasshoppers and katydids) are less so. Specialized interactions were more frequent for fern species growing in low-to-mid elevation tropical locations, that is, sites with high precipitation, high temperature, and low-variability climates. Specialization of herbivorous insects is the hallmark of tropical insect herbivores feeding on angiosperms (Forister et al., 2015). Fern-feeding insects appear to be similar in this respect.

Porto et al. conclude that the herbivore faunas of ferns are a mix of species varying in age of origin: Lepidoptera and Coleoptera that switched relatively recently from angiosperms to ferns (see Hendrix, 1980), and more ancient associations with Hemiptera and Orthoptera, supporting Cooper-Driver (1978)'s initial proposal. These results mirror the results of studies of three Hymenoptera lineages (i.e. sawflies) whose larvae feed only on ferns (Schneider, 2016). Two families of fern sawflies represent ancient associations with ferns that evolved before the origin of the angiosperms, while a third is a lineage that switched over from angiosperms after their diversification (Schneider, 2016).

Porto et al. give us a clearer picture of the degree of specificity of insects feeding on ferns today (see also Fuentes-Jacques et al., 2022). But we still have no answer as to whether ferns host fewer herbivore species than angiosperms, and whether those species are more or less specialized than the average herbivore attacking angiosperms. Answering these questions will require further comparative analyses.

By contrast, the question of whether fern defenses against herbivores are unique is becoming clearer with recent research. Extrafloral nectaries (EFNs) attract ants, and the ants, in turn, often bother, scare away or ingest insect herbivores. As a result, ants can benefit both individual angiosperm plants and fern plants (Rico-Gray & Oliveira, 2008). EFNs arose approximately contemporaneously in the two taxa, c. 132 Ma in flowering plants and c. 135 Ma in ferns (Suissa et al., 2024). Diversification followed EFN evolution in ferns, but there was a 100 million year lag between the time of EFN appearance in ferns and the surge in diversification (Suissa et al., 2024). Perhaps this diversification was driven by the colonization of ferns by herbivores from angiosperms. This possibility suggests that insect herbivores had a hand in the diversification of ferns (Suissa et al., 2024) as has been proposed for angiosperms. The specifics of how such herbivore-driven speciation would take place are still being debated (Marquis et al., 2016; Maron et al., 2019). Ants that visit EFNs may also have played a role in fern speciation. It is interesting to note that the percentage of plant species that produce EFNs is 1% in both ferns and angiosperms (Suissa et al., 2024). Thus, ferns are no less nor more likely to be defended by ants than are flowering plants.

Whether chemical defenses against herbivores are reduced in ferns remains unclear. While studies of their herbivore faunas have continued to accumulate, studies of fern secondary chemistry are still less common (Castrejón-Varela et al., 2022). Ferns have unique compounds not found in angiosperms (Wei et al., 2023), and there are compounds in angiosperms that appear to be completely absent or nearly so in ferns (Castrejón-Varela et al., 2022). Like angiosperms, ferns often have tough tissue, spines, and hairs, all shown to protect angiosperms against herbivore attack. Complete metabolic profiles, including secondary compounds, are becoming easier to acquire and analyze, but it seems that very few fern species have been profiled in this way (Wei et al., 2023). Once such data are available, it will be possible to answer if and how fern chemistry differs from that of flowering plants, and how these compounds contribute to the structuring of fern herbivore assemblages and the damage they cause. Completely overlooked is the impact of the wide diversity of leaf morphology and size, plant size, deciduousness, and habitat among fern species. These factors have been shown to influence herbivore damage in angiosperms (see review by Chaves et al., 2025). Along with the role of chemistry, the impact of these additional factors on the assembly of fern herbivore faunas awaits testing.

A very important contribution of papers by Porto et al. and Fuentes-Jacques et al. (2022) is to refocus our attention on the mysteries of interactions between the more than 12 000 extant species of ferns and their insect herbivores. Additional sampling of these interactions is especially needed in the tropics. Herbivores have been surveyed for only 3.5% of all fern species (Porto et al., 2025), and most of these are of temperate species (Schneider, 2016; Porto et al., 2025). Focused sampling of interactions, carried out in a phylogenetic context for both ferns and insects, would seem to be the most productive. It would also be revealing to compare fern clades that evolved before the appearance of angiosperms with those that evolved subsequent to their appearance.

Porto et al. point out the need to focus research and conservation efforts on tropical ferns and their insect associates, representing both the most specialized and the least studied interactions. Fern–insect interactions that occur in low-to-mid elevation tropical forests are predicted to be the most vulnerable to climate change because of their narrow climatic niches (Grinder & Wiens, 2023). The vulnerability of fern–insect interactions represents one more example of a potential loss of unique ecological interactions before the participants have been documented.

The New Phytologist Foundation remains neutral with regard to jurisdictional claims in maps and in any institutional affiliations.