Human colonization of islands has resulted in the reduction or loss of many native species, and the introduction of non‐native species, producing novel ecosystems. The impacts of these changes on mutualistic plant–animal interactions have received considerable attention, but the potential effects on some antagonistic interactions, such as seed predation, are less thoroughly understood, and often overlooked. Using three archetypal island groups—New Zealand, the Mascarenes and Hawaiʻi—we compare the taxonomic richness and functional diversity of vertebrate seed predators from prehuman and contemporary ecosystems. We scored species on several traits relevant to seed predation, then clustered species into functionally similar groups using hierarchical clustering. These archipelagos once supported between 19 and 24 species of exclusively avian seed predators (representing two to four orders) ranging from large, flightless herbivores to small, volant finches and parrots. Following human arrival, 63%–89% of these species went extinct, and between 12 and 23 non‐native seed predators were introduced. Contemporary seed predator faunas consist of between 14 and 26 species (representing six to seven orders), dominated by non‐native granivorous birds and omnivorous mammals. Our results reveal several examples in which non‐native species may be functionally similar to extinct seed predators, but most non‐native species are functionally different from extinct species, and therefore may be introducing novel seed predation pressures for insular ecosystems. Mammalian seed predators are especially functionally different from the native avian seed predators, as their teeth and widespread habitat distribution allow them to destroy a more diverse range of seeds, including the largest seeds. We highlight the need to understand how these altered seed predator communities are affecting native plant populations, particularly in the context of reduced pollination and seed dispersal. More broadly, we argue that antagonistic interactions are an integral part of any ecosystem, and therefore must be understood if we are to achieve more holistic restoration frameworks for insular ecosystems. A free Plain Language Summary can be found within the Supporting Information of this article.

Often the mutualistic roles of extinct species are inferred based on plausible assumptions, but sometimes palaeoecological evidence can overturn such inferences. We present an example from New Zealand, where it has been widely assumed that some of the largest-seeded plants were dispersed by the giant extinct herbivorous moa (Dinornithiformes). The presence of large seeds in preserved moa gizzard contents supported this hypothesis, and five slow-germinating plant species () with thick seedcoats prompted speculation about whether these plants were adapted for moa dispersal. However, we demonstrate that all these assumptions are incorrect. While large seeds were present in 48% of moa gizzards analysed, analysis of 152 moa coprolites (subfossil faeces) revealed a very fine-grained consistency unparalleled in extant herbivores, with no intact seeds larger than 3.3 mm diameter. Secondly, prolonged experimental mechanical scarification of and seeds did not reduce time to germination, providing no experimental support for the hypothesis that present-day slow germination results from the loss of scarification in moa guts. Paradoxically, although moa were New Zealand's largest native herbivores, the only seeds to survive moa gut passage intact were those of small-seeded herbs and shrubs.

Understanding marginal habitat use by invasive species is important for predicting how distributions may change under future climates. We investigated the influence of food availability and temperature on ship rat (Rattus rattus) distribution and density across a forested elevational gradient in New Zealand by measuring ship rat demographics following a beech (Nothofagaceae) mass seeding event (‘mast’) at three elevation bands (20–80, 400–500, 800–900 m asl). We tested whether declining food availability limits rat populations at the highest elevation band post-mast by experimentally increasing food abundance above baseline food availability. When our study started 4 months post-seedfall, rats at mid- and low- elevations were at high densities (11.4–16.5 ha−1). Rats at higher elevations were barely detectable, but densities peaked (9.4 ha−1) 10 months post-seedfall, with the initial increase possibly driven by immigration from lower elevations. All populations declined sharply over the next year. Supplementary feeding at high elevation increased survival, recruitment, and density of rats through winter, 16 months post-seedfall, relative to unfed grids, suggesting food limitation. However, both fed and non-fed populations declined to zero by the following spring, perhaps due to stoat (Mustela erminea) predation. Our results suggest that low food availability plays a significant role in restricting rats from cool, high elevation environments. The variation in the timing and magnitude of ship rat responses to the pulsed resource across the gradient also highlights the importance of initial population size and spatial processes as factors modulating ship rat responses to pulsed resources across a landscape.

Formulations of glyphosate-based herbicides continue to dominate the global herbicide market, while there continue to be concerns regarding the impact of this herbicide on non-target organisms. Research also indicates that the additives within certain glyphosate formulations, such as surfactants, are actually more toxic than the glyphosate active ingredient alone. Concerns arise in particular when glyphosate formulations are proposed for vegetation control in areas inhabited by rare or threatened species. Although the effect of glyphosate on birds and mammals is well studied, reptiles remain neglected in ecotoxicological studies. We investigated whether dermal exposure to two different commercial glyphosate formulations affected performance measures in the New Zealand common skink (Oligosoma polychroma). Fifty-eight skinks were each placed in a box of straw to simulate field conditions and sprayed once with Agpro Glyphosate 360, Yates Roundup Weedkiller (both at the label-specified concentrations of 144 mg glyphosate per 1 L water), or water (control). Agpro Glyphosate 360 contained ethoxylated tallow amine at a concentration of <200 g/L, while the surfactant within Yates Roundup Weedkiller was unknown. Following treatment skinks were kept in captivity and sampled for selected temperature and mass over a four-week period. Neither glyphosate formulation had a significant impact on mass. However, skinks treated with Yates Roundup Weedkiller selected significantly higher temperatures across 3 weeks following exposure. This heat-seeking behaviour could be a fever response to increase metabolism and thereby counteract physiological stress.