The major soluble carbohydrates in the desiccation-tolerant leafy liverwort Porella platyphylla (L.) Lindb. are sucrose and a homologous series of fructans including the trisaccharide 1-kestose. Exogenous glucose and fructose (10 mol m−3) did not affect the composition of the soluble carbohydrate pool. Sucrose caused an increase in the fructan pool. Sucrose also inhibited photosynthetic oxygen evolution and respiration. The fructan pool was maintained in preference to sucrose during dark starvation. Low temperature and low water potential increased the fructan pool whereas desiccation increased the proportion of high molecular weight fructan. Acid invertase activity was detected in a taxonomically diverse range of liverworts but was very low or undetectable in a range of mosses. The invertase activity from P. platyphylla was partially purified by ammonium sulphate precipitation. The reaction products of the partially purified enzyme were equimolar glucose and fructose. Kestose and higher DP fructans were not detected suggesting that, at least under the assay conditions used, the enzyme does not have sucrose[ratio ]sucrose fructosyl transferase activity. The pH optimum was 4·5–5 and the Km for sucrose was 1·7 mol m−3. Pyridoxal hydrochloride (5 mol m−3) caused 50% inhibition. The coexistence of sucrose and invertase suggests that either the invertase is inactive in vivo or is in a different subcellular compartment from sucrose. The pH response shows that it would have very low activity at cytosolic pH. A large acidic vacuole was detected in P. platyphylla leaf cells by neutral-red staining in which either invertase or sucrose could be sequestered. Rehydrating desiccated P. platyphylla for 10 min resulted in a 60% loss of extractable invertase activity. By contrast, extractable malate dehydrogenase activity increased during rehydration. Rehydrating desiccated leaves caused an increase in glucose and fructose suggesting that the sucrose pool was susceptible to invertase at this time. It is suggested that the partial inactivation of invertase during rehydration minimizes sucrose hydrolysis while membrane structure and subcellular compartmentation are re-established.

Using the outcrossing Primula farinosa and its autogamous selfing relatives P. scotica, P. scandinavica and P. stricta, we compared the fitness of light and heavy seeds. Heavy seeds germinated in greater numbers and more quickly. In competition with seedlings grown from lighter seeds, heavy seeds produced larger rosettes. In P. farinosa such seedlings went on to produce more seeds, and in two populations heavier seeds, than plants from lighter seeds. After transplantation to natural populations, seedlings of P. farinosa derived from heavy seeds produced larger rosettes, more flowers and seeds than those from lighter seeds in certain populations so that seedlings born of heavy seeds were much fitter than seedlings from lighter seeds. Average seed weight varied in inverse proportion to seed number per capsule. The autogamous species produced on average about twice as many seeds per capsule as P. farinosa. In P. scotica and P. stricta this difference appears to be due in part to assured fertilization, but this high fecundity did not cause disadvantageously light seeds. As these species produced fewer capsules per scape, their overall seed production was on average no greater than for P. farinosa. P. farinosa traded-off fitness between capsules with large seed numbers, which donated more offspring to the next generation, and those with small seed numbers, whose heavy seeds would be more likely to reproduce themselves in the next generation. We conclude that low fecundity in outcrossing species might at times be advantageous.

Accumulation of lead in the crustose lichen Acarospora smaragdula sensu lato is reported in the vicinity of an ore- processing plant where it is subjected to acidification and metal particulate fallout. A combination of light microscopy, X-ray element mapping, field emission scanning electron microscopy (FESEM) and other analytical techniques identifies Pb accumulation within specific fungal tissues derived from smelter particles (PM10s). No Pb was detected within the photobiont layer. Our studies suggest that Pb is highly mobile under the prevailing acidic conditions, and is fixed within the lichen cortex and melanized apothecia. Lead is also accumulated within the medulla and at the rock–lichen interface where it may precipitate as amorphous botryoidal encrustations on medullary hyphae and iron-rich particles. Modern FESEMs and microprobes enable analysis of minute quantities of material, and are important tools in understanding the fate of metals within lichens necessary to develop their use as predictive and sensitive bioindicators of aerial particulate contaminants. We suggest that crustose lichens, hitherto largely ignored in metal pollution studies, may be useful bioindicators of aerial particulate contaminants in polluted areas where macrolichens are absent.

Functional morphological patterns in root apices of tomato (Lycopersicon esculentum) dependent on growth, ageing and infection by the arbuscular mycorrhizal (AM) fungus Glomus mosseae and/or by the soilborne pathogenic fungus Phytophthora nicotianae var parasitica (P. parasitica) were studied. Uninfected root apices were characterized by closed, tri-layered meristems with nonreticulate nuclei; however, some apices of each treatment lost their meristematic nature, stopped growing and differentiated, becoming ‘parenchymatized’. The pathogenic fungus reduced the apex diameter and the number of mitotically active and viable apices inducing plasmolysis, cell and nucleus degeneration, and necrosis. The AM fungus, on the other hand, produced an increase in apex size and reduced the percentage of necrosis both in uninfected roots and in roots infected by P. parasitica. Thus, the AM fungus protected the apices from the pathogenic infection, allowing normal root growth. Furthermore, larger apices, which produce thicker roots, might indirectly contribute to plant protection. Increased volumes of colonizable tissues favour the spreading of the symbiont, and P. parasitica hyphae are always excluded from arbuscule-containing cells.

Extended areas of new crops to produce not only food, but also energy and materials, was one of the visions when the EU agricultural policy was reformed in 1992. However, this vision remains unfulfilled. The perennial C4 grass Miscanthus, originating from eastern Asia, was one of the crops that caught major interest as a potential biomass crop due to its high productivity even in cool northern European conditions (Beale & Long, 1995). However, even very large initial programmes on Miscanthus were conducted almost exclusively within one genotype, namely the sterile, triploid, interspecific hybrid M. × giganteus, and ran into significant problems of low first winter survival and prohibitive high costs of vegetative establishment. Thus, despite the many encouraging results on Miscanthus productivity (van der Werf et al., 1993), environmental acceptability (Christian & Riche, 1998) and harvest and storage suitability (Venturi et al., 1998), it is clear that the introduction of a new crop into agriculture is not simple, and that basic ecophysiological understanding is important to support commercial exploitation. The results in this issue from Clifton-Brown & Lewandowski (see pp. 287–294) contribute to this vital, basic understanding, revealing that low frost tolerance of M. × giganteus rhizomes is probably the cause of low winter survival in cool parts of Europe, but that within the genus Miscanthus better frost tolerance exists.

The effects of excess nitrogen (N) on the frost hardiness and freezing injury of bark and buds were tested in 2-yr-old sessile oaks (Quercus petraea) and pedunculate oaks (Q. robur) that had been grown outdoors in sand culture with normal or luxurious N supply during the growing season. Some trees from both N treatments were subjected to drought stress in summer, whereas others were adequately watered. Between January and March, whole plants were exposed to artificial freezing treatments at −25 °C of different durations (10 d, 21 d, or two periods of 4 d interrupted by an 8 d frost-free period, referred to as a freeze–thaw cycle). The frost hardiness and freezing injury of the bark were determined by two different versions of the electrolyte leakage method (calculation of an index of injury, I−25, from the relative conductivity after a fixed time; and measurement of the electrolyte leakage rate). In addition, the vitality of bark and buds was assessed visually with the 2,3,5-triphenyltetrazolium chloride (TTC) method. In comparison with the control, the oaks that had been luxuriously supplied with N exhibited distinctly higher N concentrations and lower C:N ratios in their leaves and bark, higher relative shoot increments, delayed leaf discoloration in autumn, and earlier budbreak in spring. Oaks kept outdoors during the winter were frost-hardy from December until the end of February without showing differences between species or N treatments. In comparison with the pedunculate oak, the sessile oak was more susceptible to freezing injury induced by prolonged artificial freezing or by the freeze–thaw cycle. Increased freezing injury in trees with high N supply was detected only by the TTC test of the bark of sessile oaks subjected to the freeze–thaw cycle. In all oaks, except for the pedunculate oaks grown with normal N, the freeze–thaw cycle resulted in significantly lower frost hardiness, as indicated by increased I−25 values. This treatment also led to increased freezing damage of bark and buds, as revealed by the TTC test. In drought-stressed oaks, the electrolyte leakage rate was significantly elevated after the freeze–thaw cycle. It is concluded that, in the sessile and pedunculate oaks, the water supply before frost stress and the course of the temperature in winter have a greater effect on frost hardiness and freezing damage than excess N.

The transmission of infection by many soil-borne fungal parasites of plants depends on the ability of the fungus to grow on or through soil. Progress in analysing the effects of soil physical factors on the temporal and spatial dynamics of fungal growth has been hindered by technical difficulties of quantifying fungal biomass in soil and heterogeneity in soil properties. In this paper we use a combination of a monoclonal antibody-based immunosorbent assay and microscopy to analyse the effects of soil physical properties on the spatial and temporal dynamics of colonies of the economically important fungus Rhizoctonia solani Kühn growing in two dimensions and three dimensions in a sand. Combinations of different particle-size distributions and matric potential are used to manipulate the air-filled pore volume and pore-size distribution independently of each other. Temporal dynamics are measured by the change in fungal biomass over time whereas spatial dynamics relate to fungal spread and are measured by the colony size, the rate of colony expansion and the biomass distribution within colonies. We show that the fungus spreads more than three times further over surfaces than through sand, even though the same amount of biomass is produced in each case. Pore-size distribution and air-filled pore space both affected the extent and rate of fungal spread in three dimensions within sand, with more rapid and extensive spread in a coarse sand compared with a fine sand at identical air-filled pore volume. The spread of fungal hyphae along surfaces was affected neither by differences in surface texture nor by air-filled volume, and was substantially more homogeneous than for three-dimensional spread. We argue that the relative impermeability of sand surfaces to penetration by hyphae might be influenced by the ability of the fungus to branch within a confined space rather than simply to penetrate the pores. The broader epidemiological and ecological consequences of preferential spread by parasitic and saprophytic fungi along surfaces rather than through the dense soil volume are discussed.

A simulation model for the prediction of grassland (Lolium perenne) productivity under conditions of climate change is described and validated for grass growing in the Wageningen Rhizolab, Wageningen, The Netherlands. In this work the model was used to study the impact of different management strategies on the productivity of grassland under present and increased atmospheric CO2 concentrations. In LINGRA-CC simulated key processes are light utilization, leaf formation, leaf elongation, tillering and carbon partitioning. The daily growth rate is determined by the minimum of a sink and a source term. As in a previous model (LINGRA), the potential growth of the sink depends on the mean daily temperature, and can be modified by the effects of the availability of assimilates on tillering. The growth of roots is calculated from the amount of carbohydrates the shoot is unable to utilize when the number or activity of the sinks is small (overflow hypothesis). The main difference between LINGRA and LINGRA-CC is the way the source of assimilates for growth is calculated. Assimilate production depends on intercepted radiation, and a photosynthetic light-use efficiency (LUE) calculated as a function of CO2, temperature, light intensity and the Rubisco concentration of upper leaves. Other differences are that in LINGRA-CC, the specific shoot area for new growth depends on the level of reserves. Data from two independent experiments with L. perenne swards, grown in enclosures at two levels of CO2 during 1994 and 1995, were used to calibrate and validate the model, respectively. The model predicted well the observed amounts of harvested biomass, and the dynamics of the leaf area index, tiller number and specific shoot area. LINGRA-CC was used to study the effects of different combinations of cutting interval and cutting height on biomass production, at ambient (350 μmol mol−1 CO2) and double (700 μmol mol−1 CO2 ) CO2 conditions. Under both ambient and doubled CO2, maximum biomass was produced with cuttings of leaf area index >1, and at cutting intervals of 20 and 17 d for ambient and increased CO2 environments, respectively. Under high CO2 conditions the cutting interval for maximum yield was 15% shorter than at ambient CO2. However, the gain in harvested biomass obtained by reducing the cutting interval by 3 d under high CO2 conditions was negligible.

Genomic DNA from spores of the endomycorrhizal fungus Gigaspora margarita was amplified with two sets of degenerate primers designed on conserved regions of chitin synthase polypeptides. Three chitin synthase (chs) gene portions were identified: Gimchs1, Gimchs2 and Gimchs3, encoding one class II and two class IV chitin synthases, respectively. Gimchs1 and Gimchs3 transcripts were detected by reverse transcription polymerase chain reaction (RT-PCR) during root colonization, whereas none of three mRNAs was revealed during spore germination. The potential role of the genes expressed is discussed in the light of morphogenetic and signalling events.

In studies with a laboratory isolate of the fungal pathogen Stagonospora (Septoria) nodorum three different isolates of bacteria were closely associated with the fungus. Bacteria were also closely associated with fresh isolates of S. nodorum obtained from artificially and naturally infected field material. Although a range of bacteria was isolated, only one type of bacterium was found to be associated with each isolate of S. nodorum. In co-inoculation studies with pycnidiospores of the fungus on detached leaves, some of the bacterial isolates significantly increased the pathogenicity of the fungus, particularly Xanthomonas maltophilia, Sphingobacterium multivorum, Enterobacter agglomerans and Erwinia amylovora. Evidence is presented indicating that one of the ways that the ‘helper bacteria’ may assist in the establishment of infections is by the production of lipases that were not detected in germinating fungal spores.

The potential to improve the salt tolerance of white clover (Trifolium repens L.) was evaluated in populations developed by selecting plants within the cv. Haifa which had high and low concentrations of Cl− in the shoots. Under saline conditions (40 mm NaCl), the low-Cl populations consistently maintained lower concentrations of both Na and Cl in all plant parts and produced more dry matter than the original parent cultivar (Haifa), or the high-Cl populations. Under non-saline conditions, there were no differences between populations for these characters. In young seedlings, concentrations of Cl− in the shoots of the low-Cl and high-Cl populations started to diverge after 4 d exposure to saline conditions and there were significant differences between populations after 6 d growth at 40 mm NaCl. Results using 36Cl suggested that the salt-tolerant low Cl population was better able to retain 36Cl in the roots and to limit its translocation to the leaves and petioles than the high-Cl population. Realized heritability values for shoot Cl− concentration were found to be moderate in the first cycle of selection (0·24 and 0·37 for high and low Cl− concentrations respectively), but were substantially lower (0·10 and 0·09 respectively) in the second selection cycle. We propose that it is possible to increase levels of salt tolerance in white clover by selecting for low shoot Cl− concentrations under saline conditions and that this criterion could be used at very early stages of exposure to NaCl (i.e. day 4 or 6).

Applying silicon in the form of metasilicic acid (H4SiO3) or silicic acid (H4SiO3) to Bradyrhizobium-infected, hydroponically grown cowpea seedlings resulted in a significant (P[les ]0.05) increase in the number of nodules, nodule dry matter, and nitrogen fixed on a per plant basis. Total dry matter of plants increased with silicon supply, and the differences were significant (P[les ]0.05) at the higher silicon concentrations. Cowpea plants cultured in sand were also assessed for their response to silicic acid. Nodule number and nodule mass increased with silicon supply to sand cultured plants, though nitrogen fixation was unaltered. Although silicon is not essential for growth of cowpea, it is important for nodule formation and nodule functioning in hydroponically grown plants. Consequently, data collected and conclusions drawn from earlier glasshouse experiments, which have excluded silicon from nutrient solutions, may be flawed. Future studies on nodulation and nitrogen fixation using legumes in liquid culture must therefore include silicon as a nutrient element.

Variation within the internal transcribed spacer (ITS) of the ribosomal RNA gene of 15 isolates representing seven European Armillaria species, was examined by sequencing of the PCR-amplified products. The analysis of an 744-bp region showed that the 5.8S gene appeared to be highly conserved in the 15 isolates and in other Basidiomycetes and Ascomycetes, whereas ITS1 and especially ITS2 spacers exhibited polymorphisms due to base substitutions, insertions or deletions of up to eight nucleotides. An initial dendrogram for the full sequence was drawn using cluster analysis (UPGMA), and a tree was constructed using the maximum parsimony method. Both methods indicated that the isolates could be divided into four major groups. One group, consisting of A. ectypa, was distinct from all the other species. Examination of the other groups indicated that A. tabescens and A. mellea were in a separated cluster, with a significant variation between the two isolates of the latter species. A. gallica and A. cepistipes constituted another closely related group distinguishable from A. ostoyae and A. borealis, these latter two species exhibiting the highest similarity. The results are consistent with, and discussed in regard to, the relationships estimated previously by pairing tests, morphological and physiological comparisons, as well as by restriction fragment length polymorphism of the rDNA.

The development of cereal cyst nematode (CCN; Heterodera avenae) induced syncytia in the host roots of infected resistant bread wheat (Triticum aestivum cv. AUS10894), diploid wheat (Aegilops tauschii), barley (Hordeum vulgare cv. Chebec and cv. Galleon) and in the susceptible wheat cv. Meering and barley cv. Clipper were studied over a period of 13 d. The resistance to CCN in these cereal plants is conferred by the resistance genes Cre1 in the wheat cv. AUS10894, Cre3 in A. tauschii, Ha2 in barley cv. Chebec and Ha4 in barley cv. Galleon. Anatomical observations were made on the development of the syncytia in CCN-infected wheat and barley roots, which carry each of these four sources of resistance genes. Accelerated development of the syncytia in resistant plants, especially in the barley cultivars, was observed. The sites of syncytia development in susceptible wheat and barley were also closely associated with the vascular tissues in the stele, but less so in the resistant plants. The syncytia in the infected susceptible wheat and barley were also metabolically active at day 13. By contrast, the syncytia of resistant wheat plants carrying the Cre1 or Cre3 genes remained extensively vacuolated and less metabolically active. In barley plants with the Ha2 or Ha4 genes, the syncytia appeared non-functional and in early stages of degeneration by day 13 after inoculation.

One hundred and sixty-eight sterile endophytic mycelia were isolated from roots of four Woollsia pungens (Cav.) F. Muell. (Epacridaceae) plants collected from a field site in New South Wales, Australia. All isolates formed typical ericoid mycorrhizal structures when inoculated onto roots of Vaccinium macrocarpon Ait. (Ericaceae). Microsatellite-primed PCR fingerprints generated using the primers (GTG)5 and (GACA)4 indicated that considerable genetic diversity exists within the endophyte population. It was estimated that a minimum of 43 genetically distinct mycelial genets were present in the root systems of the sampled W. pungens population, with most genets confined to individual plants. Two genets, however, were present within the root systems of two adjacent plants. While most genets were represented by less than eight isolates, three genets contained up to 41 isolates, suggesting that root system colonization by some endophytic mycelia might be extensive.

The aim of the present study was to investigate the effects of elevated CO2 on the antioxidative systems and the contents of pigments, soluble protein and lipid peroxidation in leaves of adult oaks, Quercus pubescens and Quercus ilex, grown at naturally enriched CO2 concentrations. For this purpose, a field study was conducted at two CO2 springs in Central Italy. Measurements of the pre-dawn water potentials indicated less drought stress in trees close to CO2 springs than in those grown at ambient CO2 concentrations. Most leaf constituents investigated showed significant variability between sampling dates, species and sites. The foliar contents of protein and chlorophylls were not affected in trees grown close to the CO2 vents compared with those in ambient conditions. Increases in glutathione and other soluble thiols were observed, but these responses might have been caused by a low pollution of the vents with sulphurous gases. At CO2 vents, glutathione reductase was unaffected, and superoxide dismutase activity was significantly diminished, in both species. Generally, the activities of catalase, guaiacol peroxidase and ascorbate peroxidase as well as the sum of dehydroascorbate and ascorbate were decreased in leaves from trees grown in naturally CO2-enriched environments compared with those grown at ambient CO2 concentrations. The reduction in protective enzymes did not result in increased lipid peroxidation, but increased monodehydroascorbate radical reductase and dehydroascorbate reductase activities found in leaves of Q. pubescens suggest that the smaller pool of ascorbate was subjected to higher turnover rates. These data show that changes in leaf physiology persist, even after lifetime exposure to enhanced atmospheric CO2. The results suggest that the down-regulation of protective systems, which has also previously been found in young trees or seedlings under controlled exposure to elevated CO2 concentrations, might reflect a realistic response of antioxidative defences in mature trees in a future high-CO2 world.

This study focuses on the following questions: (i) whether reductions in root:shoot ratio have a cost in terms of nutrient balance of the plant, and (ii) whether changes in resource-allocation patterns are proportional among different resources. Our approach was to analyse the variations in the allocation pattern induced by soil waterlogging. A pot experiment was conducted to analyse the effects of waterlogging on biomass, phosphorus (P) and nitrogen (N) accumulation of Paspalum dilatatum and Danthonia montevidensis, two waterlogging-tolerant grasses. When changing from oxic to anoxic conditions, a common response of these and other waterlogging-tolerant grasses is a reduction in allocation to below-ground resources. It was observed that (i) the reduction in root:shoot ratio caused by waterlogging did not have a cost in terms of capacity for nutrient uptake; (ii) resource partitioning within aerial parts was less sensitive to treatments than partitioning between roots and shoots; and (iii) biomass does not appear to be a useful currency for evaluating nutrient-allocation patterns, as the allocation of P and N was inadequately represented by biomass. The results presented here indicate that the existence of compensation mechanisms reduces the predictive value of the partition of resources for the capacity of plants to acquire resources. Data on the allocation of nutrients in relation to biomass suggest that the assumptions of independence in the allocation pattern between biomass and limiting nutrients under the effects of environmental factors can be extended.

Increased leaf phosphorus (P) concentration improved the water-use efficiency (WUE) and drought tolerance of regularly defoliated white clover plants by decreasing the rate of daily transpiration per unit leaf area in dry soil. Night transpiration was around 17% of the total daily transpiration. The improved control of transpiration in the high-P plants was associated with an increased individual leaf area and WUE that apparently resulted from net photosynthetic assimilation rate being reduced less than the reductions in the transpiration (27% vs 58%). On the other hand, greater transpiration from low-P plants was associated with poor stomatal control of transpirational loss of water, less ABA in the leaves when exposed to dry soil, and thicker and smaller leaf size compared with high-P leaves. The leaf P concentration was positively related with leaf ABA, and negatively with transpiration rates, under dry conditions (P < 0.001). However, leaf ABA was not closely related to the transpiration rate, suggesting that leaf P concentration has a greater influence than ABA on the transpiration rates.

CONTENTS

Summary 389

I. INTRODUCTION 389

II. 1. The SNARE hypothesis 393

2. SNAREs and associates 395

(a) NSF 395

(b) SNAP 395

(c) SNARE 395

(d) Rab and Rab-associated proteins 397

(e) Other interacting proteins 398

(f) Clostridial neurotoxins 399

3. New perspectives on SNARE function 400

(a) Revisiting the SNARE hypothesis 400

(b) Roles for SNARE proteins 401

III. VESICLE TRAFFICKING IN PLANTS 402

1. Identifying secretory proteins in plants 402

2. Placing syntaxins in order 403

3. Functional analysis of plant SNAREs in vesicle trafficking 404

4. Small GTPases 405

5. Searching for a vesicle–fusion complex 405

IV. REGULATION OF VESICLE TRAFFICKING IN PLANT CELLS 406

1. Hormonal control of secretion 406

(a) Auxin-evoked secretion and cell growth 406

(b) Gibberellic acid and secretion by barley aleurone 407

(c) Abscisic acid control of stomatal aperture 407

2. Coupling secretion to cytosolic-free Ca+ concentration 408

3. Stress-evoked secretion 409

(a) Mechanical stress 409

(b) Wounding and pathogen attack 409

V. CONCLUSION 410

Acknowledgements 411

References 411

Eukaryotic cells share a set of secretory pathways for the flux of membrane and protein material. In 1993, ideas about the functioning of three major proteins of the neurosecretory complex were consolidated in the SNARE hypothesis, which proposed that the interaction of these proteins provides both the specificity for vesicle targeting and the molecular machinery for fusion between vesicle and target membranes. Subsequetly, the organization, molecular mechanics and control of vesicle trafficking have become topics of intense research, and the hypothesis has evolved to accommodate new discoveries from the analysis of secretion in yeast and mammals. It is likely to be challenged again as more information comes to light about secretory processes in plants. New tools for measuring and manipulating vesicle traffic and secretion are now being used, drawing on in vivo fluorescence and capacitance recording as well as genetic engineering. These new technologies have already begun to yield details wholly unexpected from past studies. Here we focus on recent findings relating to the mechanisms of vesicle trafficking and the background to these developments, highlighting both current understanding of the molecular events of secretion and the gaps therein, as well as discussing emerging themes from work with plants.

Wetting of the upper leaf surface of Juglans regia L. and of model surfaces colonized by epiphytic micro-organisms was investigated by measuring contact angles of aqueous solutions buffered at different pH values. During June to October 1995, contact angles of aqueous solutions on the leaf surface of J. regia decreased by angles ranging from 12° (low pH values) to 25° at high pH values. At the end of this vegetation period, wetting was strongly dependent on pH showing significantly lower contact angles with alkaline solutions (pH 9·0) than with acidic solutions (pH 3·0). Contact angle titration measured angles on the leaf surface as a function of the pH of buffered aqueous solutions, covering a pH range from 3·0 to 11·0. Titration curves revealed inflection points around 7·5, indicating the existence of ionizable carboxylic groups at the interface of the phylloplane. Altered leaf-surface wetting properties observed on the intact leaf surface could be simulated in model experiments by measuring contact angles on artificial surfaces colonized by Pseudomonas fluorescens and by epiphytic micro-organisms isolated from the phylloplane of J. regia. Strong evidence is provided that interfacial carboxylic groups derive from epiphytic micro-organisms present on the phylloplane. Results suggest that the age-dependent increase in, and pH dependence of, wetting as leaves mature are related to the presence of epiphytic micro-organisms on the phylloplane. Ecological consequences of increased leaf-surface wetting, concerning the structure of the leaf surface as a microhabitat for epiphytic micro-organisms, are discussed.