Steven D. Johnson’s research while affiliated with University of KwaZulu-Natal and other places

Pollination is an essential process for the reproduction of most plants and usually involves animals as agents of pollen dispersal. Pollination biology in the broad sense is therefore relevant to many different scientific fields. Here we examine how pollination biology in South Africa has contributed to our understanding of evolutionary biology, fundamental and applied ecology and agriculture. Due to its high levels of biodiversity and relatively intact ecosystems, South Africa has unique geographical advantages for research involving pollination systems. Some of the highlights of evolutionary work include studies of the functions of floral structure, signals and rewards, including various kind of floral deception, elucidation of convergent floral evolution in plants specialized for pollination by particular animal groups, and advances in our understanding of coevolu-tion and pollinator-driven plant speciation. In terms of ecology, there has been impactful research on the ecological dependence of plants on pollinators for seed production and the influences of population and community contexts on pollination processes, as well as applied applications in terms of risks of mutualism failure and the use of plant reproductive traits for predicting the spread of invasive species. In agriculture, there have been groundbreaking studies of the biology of honeybees and estimates of the value of natural habitats for pollinator services. Studies of pollination biology have proven highly valuable in advancing understanding in a wide range of fields in biology globally, and work in South Africa has contributed significantly, particularly over the past five decades.

Species can be difficult to circumscribe using morphology alone, and additional data can thus be useful for resolving challenging species complexes. Here we provide a case study of a South African orchid species complex which we resolved by analysing morphology, floral scent chemistry and molecular data. Disa crassicornis was considered a highly variable species, but our phylogenetic analysis of samples from 15 populations showed that, as currently delimited, it is a paraphyletic entity with two accepted species, D. thodei and D. zuluensis nested in the lineage. We measured multiple morphological traits of 100 herbarium accessions and used multivariate analysis and bivariate scatter plots that identify morphological discontinuities among plants in this lineage. We identified three distinct forms among plants considered to be D. crassicornis . These forms also differed markedly in floral scent chemistry, notably the absence of the phenylpropanoid compound eugenol and its derivatives in the lowland form. We describe the lowland form with a broad platform‐like labellum as a new species, D. campestris , and restrict the name D. crassicornis to a form with broad sepals and a narrow pendant labellum that is found in the Amatola and Witteberg Mountains of the Eastern Cape. We also reinstate an earlier name D. jacottetiae for a montane form with narrow recurved perianth parts that is found at high elevation in the Drakensberg. Disa thodei and D. zuluensis are retained as distinct taxa. This study shows the value of integrative taxonomy for resolving difficult species complexes. The resolution of this species complex assists in efforts to conserve these rare taxa and also provides a sound basis for the study of ecology‐driven diversification in this lineage.

Background and Aims Intraspecific variation in floral traits can reflect shifts among different pollinator niches. We investigated whether a geographical mosaic of pollinator niches can explain divergence of flower colour and morphology among populations of Disa porrecta, a food-deceptive orchid in South Africa Methods We used field observations and pollen load analysis to establish pollinators of D. porrecta and measured spectral reflectance, morphology and scent emissions in populations. We used reciprocal translocation experiments and pairs of artificial inflorescences to establish pollinator foraging preferences. A phylogenetic analysis was used to assess relationships among populations of D. porrecta and related taxa. Results We documented two colour forms of D. porrecta that differ in pollinator assemblages. The orange-flowered form in the east of the distribution is pollinated solely by a wide-ranging nymphalid butterfly species. The pink-flowered form in the westernmost part of the distribution is pollinated primarily by a long-proboscid fly species. The orchid’s floral spectral reflectance and morphology closely matches that of most nectar plants used by these pollinators, implying a potential role for mimicry. In reciprocal translocations, butterflies showed strong preference for the orange-flowered form of the orchid at sites in the east, while both long-proboscid flies and butterflies showed a preference for the pink-flowered form at sites in the west. These preferences were also evident in experiments involving pairs of artificial inflorescences that varied only in colour. Phylogenetic analysis showed that the two colour forms of Disa porrecta share an immediate common ancestor, but the direction of the colour shift could not be reliably reconstructed. Conclusion This study identifies a role for a geographical mosaic of nectar plants and pollinators in the divergence of floral traits in D. porrecta. Differences in flower colour among populations of this orchid species are strongly correlated with pollinator foraging preferences.

Societal Impact Statement Seed dispersal is critical for the establishment and persistence of populations of most plant species. We investigated the seed dispersal biology of an African melon, Cucumis humifructus, which is closely related to cultivated cucumbers and watermelons but differs in that it buries its fruits deep underground. The fruits are located by antbears using olfaction; the antbears consume the fruit pulp and the seeds germinate in their faeces. Cucumis humifructus has become very rare in some parts of Africa and is vulnerable on account of its annual lifecycle combined with its dependence on a declining mammal species for establishing each new generation. Summary Seed dispersal mutualisms between plants and animals are seldom specialized at the species level. We investigated the highly unusual case of the dispersal of seeds of a melon Cucumis humifructus by the antbear (Orycteropus afer), a myrmecophagous African mammal. This annual plant buries its fruits c. 20 cm underground, a depth from which seedling emergence is impossible. We asked why the fruits are buried and how they are located by animals. We investigated the seed dispersal system of C. humifructus in central Namibia using camera traps, seed germination experiments and analyses of animal faeces. Coupled gas chromatography and mass spectrometry (GCMS) was used to analyse the chemistry of fruit scent. Naturally‐buried C. humifructus fruits were excavated solely by antbears. Antbear faecal pellets often contained intact C. humifructus seeds. All of the C. humifructus plants that we examined originated in antbear pellets. Fruits placed experimentally above ground or buried shallowly were eaten by porcupines which act as seed predators as they destroy soft melon seeds with their chewing teeth. Fruits of C. humifructus emit scent with a chemical profile distinct from that of related melon species. We argue that deep burial of fruits is a strategy that C. humifructus uses to escape from mammalian seed predators, particularly porcupines. This study highlights the potential role of escape from antagonistic interactions in the evolution of specialized seed dispersal mutualisms.

Ecological niche shifts are a key driver of phenotypic divergence and contribute to isolating barriers among lineages. For many groups of organisms, the history of these shifts and associated trait–environment correlations are well-documented at the macroevolutionary level. However, the processes that generate these patterns are initiated below the species level, often by the formation of ecotypes in contrasting environments. Here, I review the evidence in plants for ‘pollination ecotypes’ as microevolutionary responses to environmental gradients in pollinator availability. Pollinators are critical for population establishment and persistence in most species, thereby forming part of their fundamental niche. Novel floral trait combinations allow species to exploit particular pollination opportunities in local habitats and evolve primarily through sexual selection due to their effects on mating success. I examine selected case studies on the evolution of pollination ecotypes, including self-pollinating forms, and use these to illustrate challenging practical and conceptual issues. These issues include the paucity of reliable natural history data, the problem of implementing and interpreting reciprocal translocation experiments, and establishing criteria for when allopatric ecotypes should be considered species.

Immobility of flowering plants requires them to engage pollen vectors to outcross, introducing considerable inefficiency in the conversion of pollen production into sired seeds. Whether inefficiencies influence the evolution of the relative resource allocation to female and male functions has been debated for more than 40 years. Whereas early models suggested no effect, negative interspecific relations of mean pollen production and pollen : ovule ratios to the proportion of removed pollen that is exported to stigmas (pollen‐transfer efficiency) indicate otherwise. Here, we consider theoretically a key condition that determines whether the efficiencies of processes (first derivative of process output with respect to input) affect the evolutionarily stable sex (ESS) allocation. No effect arises if all individuals experience the same efficiency. By contrast, a decline in process efficiency with increasing allocation (diminishing returns) generally reduces the ESS male allocation for a population. Furthermore, differences in the allocation dependence of efficiencies (and hence the ESS sex allocation) among populations/species create a negative relation of realised efficiency to male allocation among species, like that observed empirically. Diminishing returns arise for various processes that affect siring (e.g. pollen export and local pollen competition to fertilise ovules), which may differ in their relative influence on sex allocation among species.

Short‐tongued saprophilous flies are a diverse but under‐appreciated group of pollinators, which are particularly important for flowers in the subtribe Stapeliinae (Apocynaceae‐Asclepiadoideae‐Ceropegieae). This clade of plants is characterised by repeated shifts between tubular kettle‐trap and open non‐trap flowers and chemical mimicry (of decomposing substrates or dying insects) to attract specific fly pollinators. The biology of most Stapeliinae, particularly those with non‐trap flowers, remains poorly studied, hampering our understanding of the mechanisms driving diversification in this plant group. We examined the pollination biology and floral traits of Ceropegia pulchellior , a South African endemic with non‐trap flowers confined to Natal Group Sandstone on the Durban escarpment. Observations showed that flowers are visited exclusively by flies which forage on small amounts of nectar in the corona cavities. Flowers were pollinated primarily by lauxaniid flies in the genus Cestrotus , although several muscid fly species also contribute. Pollinator exclusion experiments confirmed that the plants depend on these flies for reproduction. Reproductive success was low (fruit set never exceeding 8% of flowers). Pollen transfer efficiency was relatively high but variable across the flowering period. GC–MS analysis of floral scent revealed that the foetid odour is dominated by aliphatic acids and p ‐cresol with small amounts of indole, supporting the assumption that C. pulchellior mimics decaying substrates to selectively attract detritus‐feeding (saprophagous) flies as pollinators. Analysis of spectral reflectance of flowers indicates that flower colours, when viewed by the fly pollinators, are not chromatically distinct from the habitat background, suggesting that flowers rely on olfactory rather than visual signals to attract pollinators. This study contributes to the growing awareness of the complexity of pollination systems in the Stapeliinae, in terms of the wide diversity of fly taxa involved, the intricate pollen transfer mechanisms, and the unusual floral scent chemistry associated with mimicry of oviposition substrates and food sources.

Identifying the factors that contribute to reproductive isolation among closely-related species is key for understanding the diversification of lineages. In this study, we investigate the strength of premating and postmating reproductive isolation barriers between Disa ferruginea and Disa gladioliflora, a pair of closely-related species, often found co-flowering in sympatry. Both species are non-rewarding and rely on mimicry of different rewarding model flowers for attraction of pollinators. We constructed abiotic niche models for different forms of each taxon to measure ecogeographic isolation. Using experimental arrays in sympatry, we recorded pollinator transitions to measure ethological isolation. We performed hand pollinations to measure postpollination isolation. We found strong, but not complete, premating isolation associated with abiotic niches and absolute ethological isolation based on pollinator preferences in sympatry. Pollinator preferences among the orchids could be explained largely by flower colour (orange in D. ferruginea and pink in D. gladioliflora) which matches that of the pollinator food plants. Post-mating barriers were weak as the species were found to be inter-fertile. Coexistence in the orchid species pair is due mainly to ethological reproductive isolation arising from flower colour differences resulting from mimicry of different rewarding plants. These results highlight the importance of signalling traits for ethological isolation of closely-related species with specialized pollination systems.

Premise Volatile emissions from flowers and fruits play a key role in signalling to animals responsible for pollination and seed dispersal. Here, we investigated the pollination biology and chemical ecology of reproduction in Apodolirion buchananii , an African amaryllid that flowers in a leafless state soon after grassland vegetation is burnt in the dry late‐winter season. Methods Pollinators were identified through field collection and pollen loads were quantified. Floral traits including spectral reflectance and scent chemistry were documented. Bioassays using cup traps were used to test the function of floral volatiles. Fruiting biology was investigated using controlled hand‐pollination experiments and chemical analysis of fruit scent. Seed germination was scored in greenhouse trials. Seed dispersal was monitored using observations and camera trapping. Results The sweetly scented white flowers of A. buchananii are pollen‐rewarding and pollinated mainly by a diverse assemblage of bees. Cup‐trap experiments demonstrated that pollinators are attracted to phenylacetaldehyde, the dominant volatile in the floral scent. Plants are shown to be self‐incompatible, and the fleshy fruits were found to emerge from the soil six months after pollination during the peak of the summer rains. Fruits emit a diverse blend of aliphatic and aromatic esters and contain large fleshy recalcitrant seeds which germinate within days of fruits splitting open. Seed dispersal by ants was recorded. Conclusions This first account of the reproductive biology of a species in the genus Apodolirion highlights an outcrossing mating system involving bees attracted to color and scent as well as the unusual fruiting biology and ant‐mediated system of seed dispersal.

Premise Heterostyly in plants promotes pollen transfer between floral morphs, because female and male sex organs are located at roughly reciprocal heights within the flowers of each morph. Reciprocity indices, which assess the one‐dimensional variation in the height of sex organs, are used to define the phenotypic structure of heterostyly in plant populations and to make inferences about selection. Other reciprocal stylar polymorphisms (e.g., enantiostyly) may function in a similar manner to heterostyly. In‐depth assessment of their potential fit with pollinators requires accounting for the multidimensional variation in the location of sex organs. Methods and Results We have adapted the existing reciprocity indices used for heterostylous plant populations to incorporate multidimensional data. We illustrate the computation of the adapted and original indices in the freely available R package FlowerMate. Conclusions FlowerMate provides fast computation of reliable indices to facilitate understanding of the evolution and function of the full diversity of reciprocal polymorphisms.