Summary1 Angiosperm trees often dominate forests growing in resource-rich habitats, whereas conifers are generally restricted to less productive habitats. It has been suggested that conifers may be displaced by angiosperms except where competition is less intense, because conifer seedlings are inherently slow growing, and are outpaced by faster-growing angiosperm species. Here we investigate whether competition with ferns and deeply shading trees also contributes to a failure of conifers to regenerate in resource-rich habitats. 2 We examined how changes in soil nutrient availability and drainage affected vegetation along the retrogressive stages of a soil chronosequence in southern New Zealand. Vegetation composition shifted from angiosperm-tree dominance on 'recent' alluvial terraces (< 24 ky), via coniferous-tree dominance on older marine terraces (79-121 ky), to coniferous-shrub dominance on the oldest marine terraces (291 ky). Soil drainage deteriorated along the sequence, and N : P leaves and N : P soil indicate increasing Plimitation. Conifer species appear to be adapted to persistence on infertile and poorly drained soils. 3 The floor of the relatively fertile alluvial forests was deeply shaded ( ∼ 1% light transmission) by dense groves of tree-ferns and ground-ferns, and by large-leaved subcanopy trees. Few seedlings of any type were found on the forest floor, even in tree-fall gaps, and establishment was restricted to rotting logs and tree-fern trunks. Angiosperms were particularly successful at colonizing these raised surfaces. 4 Less shade was cast by the conifer-dominated forests on infertile marine terraces ( ∼ 5% light transmission), which lacked tall ferns. There were many opportunities for conifer establishment, with high seedling densities recorded on the forest floor and on logs. By contrast, angiosperm seedlings were mainly restricted to logs. 5 Our results suggest that several mechanisms act in concert to reduce regeneration opportunities for conifers in productive habitats. In particular, we suggest that tall ferns and deep shade are responsible for a restriction of regeneration opportunities in relatively productive forests in New Zealand, diminishing the opportunities for conifers to escape the competitive effects of fast-growing angiosperms. Thus 'crocodiles' may alter the outcome of the race between 'hares' and 'tortoises'.

Invasive species are frequently the target of eradication or control programmes to mitigate their impacts. However, manipulating single species in isolation can lead to unexpected consequences for other species, with outcomes such as mesopredator release demonstrated both theoretically and empirically in vertebrate assemblages with at least two trophic levels. Less is known about the consequences of species removal in more complex assemblages where a greater number of interacting invaders increases the potential for selective species removal to result in unexpected changes in community structure. Using a replicated Before-After Control-Impact field experiment with a four-species assemblage of invasive mammals we show that species interactions in the community are dominated by competition rather than predation. There was no measurable response of two mesopredators (rats and mice) following control of the top predator (stoats), but there was competitive release of rats following removal of a herbivore (possums), and competitive release of mice following removal of rats.

We analyzed seed production of mountain beech (Nothofagus solandri var. cliffortioides) forest along an elevational gradient in New Zealand from 1020 to 1370 m (treeline) for the years 1973–2002. We used seed production data from nine elevations and a site‐ and species‐specific net carbon (C) availability model from two elevations (1050 m and 1340 m) to examine how three variables (temperature, soil moisture, and net C availability) during three key periods (resource priming, flowering primordia development, and flowering) influence seed production. These three strongly interrelated variables have all been considered determinants of seed production but have not previously been tested together in a single analysis. Seed production increased over the 30‐year period, with the greatest increases at high elevations; this increase was driven by a greater frequency of intermediate‐sized seeding years. We then determined how temperature, soil moisture, and net C availability determined seeding, and examined whether temporal trends in the seeding data could be linked to similar temporal trends in temperature, soil moisture, or net C availability. High seed production was related to cool summers with high soil moisture during resource priming, warm summers during flower primordia development, and low net C availability during flowering. Positive temporal trends in temperatures during the period of flower primordia development accounted for the increase in seed production, suggesting that increasing temperatures are promoting more frequent seed production at high elevations.

Summary 1.The effect that the proximity of habitats containing essential resources has on animal abundance at large spatial scales is called landscape complementation. Landscape complementation can influence interaction between large herbivores and their food resources where the proximity of habitats containing essential resources constrains their foraging or demographic efficiency. 2. We tested the effect that the proximity of a thermal refuge (riverine woodlands) had on interaction between wild pigs and their food resources (pasture). The numerical response of pigs ( Sus scrofa L.) (estimated as r quarter − 1 ) to pasture biomass was contrasted between four sites that were progressively more isolated from a major floodplain containing extensive areas of riverine woodland. To test whether proximity to riverine woodlands affected the numerical response of pigs through a constraint on foraging efficiency, we contrasted the numerical response of pigs between the four areas as pasture biomass declined. To test whether pigs exploited riverine woodlands primarily as a thermal refuge, we contrasted the numerical response of pigs between the four areas in different seasons. 3. We found that although pasture biomass was similar in the four areas, r was lower than expected for given pasture biomass on the two areas further away from riverine woodlands. We also found that while the effect that proximity to riverine woodlands had on the numerical response of pigs became more pronounced when pasture biomass was low, it was not significantly affected by season. 4. These results suggest that the need to access riverine woodlands compromises the foraging efficiency of wild pigs when the distance to this habitat is relatively high, but that the need to access this habitat may not be purely related to thermoregulation. 5. We developed a simple mechanistic model that allows the effects of landscape complementation on herbivore foraging and demographic efficiency to be estimated, and used the model to predict the effect that proximity to riverine woodlands would have on variation in pig density. The model suggests that wild pigs cannot persist in areas more then 10 km from extensive riverine woodlands, unless those areas are periodically recolonized. This suggests that at the margin of their range around inland river systems, given locations that can be occupied by wild pigs will vary temporally between being sources, pseudosinks and sinks.