Within Apocynaceae, interactions with pollinators are highly structured both phylogenetically and biogeographically. Variation in transition rates between pollination systems suggest constraints on their evolution, whereas regional differences point to environmental effects such as filtering of certain pollinators from habitats. This is the most extensive analysis of its type so far attempted and gives important insights into the diversity and evolution of pollination systems in large clades.

Grasshoppers constitute a significant proportion of invertebrate diversity in grasslands, but little is known about how fire and grazing, two of the main drivers of grassland dynamics, influence them. Here, we determine how these disturbances influence grasshopper diversity in remnant grassland ecological networks (ENs) among commercial forestry compartments in Afromontane grassland, South Africa. Sites differed in fire frequency (annual vs. longer rotation burning), time since last fire (<12 months vs. >12 months ago), and presence or absence of domestic cattle grazing. Reference sites were in an adjacent protected area (PA). We also recorded bare ground, vegetation cover, and vegetation height. Grasshopper abundance significantly benefitted from annual burning, cattle grazing, short grass, and when the last fire was <12 months prior to sampling. Fire frequency was the only disturbance with a significant effect on grasshopper species composition, but without significantly affecting species richness. In the PA, heterogeneity of dispersion was smaller in annually burned firebreaks than in sites with longer fire‐return intervals, which points to homogenisation of resident grasshopper assemblages. Although most grasshopper species favoured recently burned or grazed grassland, some preferred low or no disturbance. This illustrates a considerable range of habitat preferences in the grasshopper assemblage. Management recommendation: Burning and grazing at moderate levels should be integral to local landscape management, as these impacts benefit grasshoppers in grassland ENs. Some areas should be left undisturbed to create a mosaic of different successional stages at the landscape spatial scale, providing diverse habitat for a wide range of species.

No abstract

Abbreviations DBERG = Drakensberg Foothill Moist Grassland; DEM = Dynamic Equilibrium Model; IDH = Intermediate Disturbance Hypothesis; KZN = KwaZulu-Natal; MBELT = Midlands Mistbelt Grassland; RB = recently burned i.e. burned <12 months prior to sampling; UB = unburned i.e. not burned <12 months prior to sampling. Nomenclature Mucina et al. (2006) for grassland types; Plants of southern Africa (http://posa.sanbi.org/) for vascular plants AbstractAims: Natural disturbances drive plant assemblages in Afromontane grassland.Our aim is to assess how intensity of cattle grazing and time since last fire influence plant species richness, dominance and assemblage composition.Location: Central Midlands, KwaZulu-Natal province, South Africa.Methods: Sites differed in grazing intensity (light, moderate and heavy) and time since last fire: recently burned (RB) vs unburned (UB). We sampled plants in 24 vegetation quadrats (1-m 2 ) randomly located within each site (~1000 m 2 ). We used linear regression models for species richness and dominance, IndVal to identify indicator species and CCA for assemblage composition. All analyses were conducted at quadrat and site level.Results: Species richness was significantly influenced by time since last fire (RB > UB), grazing intensity (light < moderate > heavy) and the fire-grazing interaction at quadrat level. Dominance reciprocated the response of species richness to time since last fire (UB > RB), but did not respond consistently to grazing intensity (light > moderate > or < heavy). Plant assemblage composition was significantly influenced by time since last fire, grazing intensity and vegetation structure (live vegetation cover and vegetation height). Indicator species of species-poor, heavily grazed sites included alien species, such as Cirsium vulgare and Conyza bonariensis.Conclusion: Response of species richness to grazing intensity conforms to the predictions of the intermediate disturbance hypothesis. Burning and moderate grazing should be part of grassland management. To prevent species loss and influx of alien species, heavy grazing should be minimized. As development of grassland ecological networks in a timber matrix is being pioneered in South Africa, these findings indicate a way forward for the management of similar systems elsewhere in the world.

No abstract