METSÄNTUTKIMUSLAITOKSEN TIEDONANTOJA 784, 2000 FINNISH FOREST RESEARCH INSTITUTE, RESEARCH PAPERS 784,2000 Heavy metal resistance of two boreal dwarf shrubs in a polluted environment Satu Monni VANTAAN TUTKIMUSKESKUS-VANTAA RESEARCH CENTRE METSÄNTUTKIMUSLAITOKSEN TIEDONANTOJA 784, 2000 FINNISH FOREST RESEARCH INSTITUTE, RESEARCH PAPERS 784, 2000 Heavy metal resistance of two boreal dwarf shrubs in a polluted environment Satu Monni Finnish Forest Research Institute Vantaa Research Centre Academic dissertation in Systematic Biology Faculty of Science University of Helsinki To be presented, with the permission of the Faculty of Science of the University of Helsinki, for public criticism in Lecture Hall of Department of Ecology and Systematics (Unioninkatu 44, Helsinki) on October 27th, 2000, at 12 o'clock noon. Helsinki, 2000 Supervisor: Dr. Ahti Mäkinen University of Helsinki, Finland Department of Ecology and Systematics Rewievers: Professor Sirkku Manninen University of Oulu, Finland Department of Biology Professor Lauri Kärenlampi University of Kuopio, Finland Department of Ecology and Environmental Science Opponent: Associate Professor Nicholas M. Dickinson Liverpool John Moores University, U.K. School of Biological and Earth Sciences Hakapaino Oy, Helsinki 2000 ISBN 951-40-1749-8 ISSN 0358-4283 ACKNOWLEDGEMENTS This work was carried out at the Vantaa Research Centre of the Finnish Forest Research Institute. I would like to thank Professor Eero Paavilainen, the former head of the Vantaa Research Centre for providing me the excellent working facilities and support of Finnish Forest Research Institute. The work was funded by the large amount of foundations; Biologian Seura Vanamo, Suomen Luonnonsuojelun Säätiö, Jenny ja Antti Wihurin Rahasto, Alfred Kordelinin yleinen edistys- ja sivistysrahasto, Maj ja Tor Nesslingin Säätiö, Suomen Metsätieteellinen Seura and Nordic Academy for Advanced Study (NorFA) which are greatly acknowledged. I want to thank my official supervisor Ahti Mäkinen for his continuous support and encouragement for the scientific work. Professor Lauri Kärenlampi and Professor Sirkku Manninen are greatly acknowledged for reviewing this thesis and giving constructive comments on it. I also want to thank Professors Timo Koponen and Heikki Hänninen from the Department of Ecology and Systematics for commenting the earlier version of the thesis and Professor Timo Koponen for handling the practical work concerning dissertation. I want to express the sincere gratitude to all of the co-authors for the pleasant co-operation. I want to thank Maija for a lot of help, introducing me the heavy metal research in Metla and the nice co-operation in greenhouse and field during these past four years. I also want to thank for several interesting discussions about heavy metal resistance and dwarf shrubs and especially for the help in the statistical analysis, which would have been hard without your good knowledge in statistics. I also want to thank Christian from the Norwegian Crop Research Institute, for the enormous support and encouragement from the beginning of my thesis, introducing me a lot of literature about Empetrum and providing valuable discussions. Without your support and enthusiasm this thesis would not be as it is. I am very grateful for the expertise of the collaborators in Germany and Norway, where this thesis has partly been done. The months in Germany and Norway were one of the best moments in research. I want to thank especially Dr. Ingrid Kottke from the University of Tubingen for inviting me to Tubingen, introducing me the techniques of electron microscopy and the endless talks about the science. In spite of work I also want to thank Dr. Kottke for introducing me the excellent classical concerts in Tubingen due to our common interest in music. The whole Department of Special Botany and Mycology is thanked for the help in daily life and nice atmosphere. I want to express many thanks to Dr. Heike Bucking from the University of Bremen for introducing me the interesting world of EDXS and giving me so much help for the article of electron microscopy. I enjoyed the encouranging but joyful atmosphere in UFT of which the whole group is acknowledged. I also want to thank Dr. Elisabeth Magel and Professor Hampp at the Department of Physiological Ecology of Plants, in Tubingen, for good supervision of organic acid analysis. Olavi, Espen and all the other smiling faces at the Department of Plant Physiology and Microbiology, in Tromso, are greatly acknowledged for a lot of help and making the days at the University pleasant. Thank you also for being responsible in getting finally ABA extracted from Empetrum. In Metla the support from enormous amount of people have made the work much easier. First I want to thank Heljä-Sisko. Maija. Christian, Tiina N.. Oili, John and Ilkka V. for the field trips to Haijavalta and the valuable discussions about the research at Harjavalta area. Secretaries Riitta H.. Pirkko R. and Annikki are acknowledged for doing the paperwork. Maarit R.. Pirkko R.. Leppis and Maija R have done chemical analyses at the 'central laboratory' of Metla and Ilkka T. has done layouts for posters. Rauski offered her helping hand to prepare the samples when the time was running out! In Ruotsinkylä greenhouse especially Lauri, Satu R.. Satu S. and Kaarina have been taking care of the plants and practical work and Maarit K. has advised to perform the water potential measurements. John has revised the English of all the five publications and this thesis. I also want to thank Jarkko K. from the University of Helsinki, Department of Forest Ecology, for using a lot of time for supervising the photosvnthetic measurements and Jyrki Juhanoja from the University of Helsinki. Department of Electron Microcopy, for valuable advice during the electron microscopical study. In spite of work I have also had pleasant moments in Metla which have made the work much more fun. Thank you 'soil group' Outi, Laura, Taina, Päivi, Satu and Oili for the great moments in and outside the work and sharing the ups and downs during the years of the thesis. I also want to thank Liisa U., Tiina N., Leila, Maija, Anna-Maija and Anne- Marie for the refreshening discussions about the research work or activities outside the work. It has also been pleasant to have 'old' good friends Minna and Tuija in Metla to get thoughts away from science in coffee breaks. I want to thank my current bosses Tomas and Mika from Electrowatt-Ekono Oy for the support and letting me to finish my PhD in spite of a lot of work in environmental consulting business. I want to give my warmest thanks to my friends and family. Thank you father and mother for believing me and taking care of my welfare. Thank you father for joining me in the field when nobody else was able to. I also want to give my dearest thanks for my sisters Outi and Suvi for the endless support and taking part to the practical things concerning thesis. Vantaa, October, 2000 TABLE OF CONTENTS LIST OF ORIGINAL PUBLICATIONS 1. INTRODUCTION 2 1.1. Mechanisms of plants to resist elevated heavy metal concentrations in the SUBSTRATE 2 1.2. Toxicity of Cu and Ni and parameters used to evaluate the responses of PLANTS TO HEAVY METALS 3 1.2.1. Toxic Cu and Ni concentrations for plants 3 1.2.2. Elongation and biomass 4 1.2.3. Physiological measurements 4 1.2.4. Electron microscopical investigations 5 1.3. Ecology and special characteristics of dwarf shrubs colonising heavy metal CONTAMINATED AREAS 5 1.3.1. Ecology of Empetrum nigrum 5 1.3.2. Ecology of Calluna vulgaris 7 1.3.3. Dwarf shrubs in heavy metal polluted areas 8 1.4. Aims 10 2. MATERIALS AND METHODS 11 2.1. Greenhouse experiments 11 2.2. The study sites near Harjavalta Cu-Ni smelter 13 2.3. Chemical analyses 15 2.4. Physiological measurements 15 2.5. Electron microscopy 17 2.6. Statistical analysis 17 3. RESULTS 18 3.1. Chemical composition of E. nigrum and C. vulgaris in response to element APPLICATIONS (I, 11, III) 18 3.1.1. Accumulation of Cu and/or Ni in E. nigrum (I) 20 3.1.2. Accumulation ofCu, Ni, Pb, Zn, Fe, Cd, Mn , Mg, Ca, P, K, Nin E. nigrum (III) 20 3.1.3. Accumulation of Cu in C. vulgaris (II) 20 3.2. ECOPHYSIOLOGICAL RESPONSES OF E. NIGRUM TO METALS IN THE GREENHOUSE AND FIELD (111. IV) 20 3.2.1. Photosynthesis ofE. nigrum (111, IV) 20 3.2.2. Stem water potential and ABA ofE. nigrum (111, IV) 21 3.2.3. Organic acid contents of E. nigrum in the field (IV) 22 3.3. ULTRASTRUCTURAL ELEMENT LOCALIZATION OF E. NIGRUM IN THE FIELD (V) 22 3.4. Growth of E. nigrum and C. vulgaris in response to Cu and/or Ni applications (I. II) 23 3.4.1. The elongation and biomass ofE. nigrum in response to Cu and/or Ni (1) 23 3.4.2. The elongation and biomass of C. vulgaris in response to Cu (11) 24 4. DISCUSSION 25 4.1. Heavy metal accumulation in C. vulgaris and E. nigrum in controlled CONDITIONS 25 4.1.1. Measured concentrations 25 4.1.2. Accumulation pattern 26 4.2. ECOPHYSIOLOGICAL RESPONSES OF£. NIGRUM TO ELEMENTS IN GREENHOUSE AND FIELD CONDITIONS 27 4.2.1 Photosynthesis of E. nigrum 27 4.2.2. Water stress of E. nigrum 28 4.2.3. Organic acid contents ofE. nigrum 29 4.3. ULTRASTRUCTURAL LOCALIZATION OF ELEMENTS IN E. NIGRUM 29 4.4. The morphological responses of C. vulgaris and E. nigrum to Cu and/or Ni in CONTROLLED CONDITIONS 31 4.5. THE HEAVY METAL RESISTANCE OF C. VULGARIS ANDE. NIGRUM 32 4.5.1. Mechanisms 32 4.5.2. The estimation ofhea\'y metal resistance in the greenhouse and field 35 5. SUMMARY AND CONCLUSIONS 37 6. REFERENCES 39 LIST OF ORIGINAL PUBLICATIONS This thesis is based on the following articles, which in the text will be referred to by their Roman numerals. I Monni, S., Salemaa, M., Millar, N., 2000. The tolerance of Empetrum nigrum to copper and nickel. Environmental Pollution 109. 221-229. II Monni, S.. Salemaa, M.. White, C., Tuittila, E., Huopalainen, M., 2000. Copper resistance of Calluna vulgaris originating from the pollution gradient of a Cu-Ni smelter, in southwest Finland. Environmental Pollution 109, 211-219. 111 Monni, S., Uhlig, C., Junttila, 0.. Hansen. E.. Hynynen. J. 2000. Chemical com position and ecophysiological responses of Empetrum nigrum to aboveground ele ment application. Environmental Pollution (in press). IV Monni. S.. Uhlig, C.. Hansen, E.. Magel. E.. 2000. Ecophysiological responses of Empetrum nigrum to heavy metal pollution. Environmental Pollution 112 (in press). V Monni, S., Bucking. H.. Kottke. 1.. 2000. Ultrastructural element localization by EDXS in Empetrum nigrum. Manuscript. 2 1. INTRODUCTION 1.1. Mechanisms of plants to resist elevated heavy metal concentrations in the substrate Heavy metal resistance is species- and metal-specific (Baker and Walker. 1990), and it can be achieved by two strategies: tolerance and avoidance (Baker. 1981, 1987). Tolerance is the ability of a plant to cope with metals that have excessively accumulated in the plant parts. Avoidance is the ability to prevent excessive metal uptake into plant parts. Tolerance mechanisms of higher plants include the production of intracellular metal-binding com pounds (e.g. organic acids or proteins), alteration of the metal compartmentation patterns (e.g. translocation of metals to older plant parts or different cell organelles), changes in cel lular metabolism (increased enzyme synthesis) or alterations to the membrane structure. Avoidance can be achieved through changes in the metal-binding capacity of the cell wall or increased exudation of metal-chelating substances (Figure 1) (Verkleij and Schat. 1990; Ko chian and Garvin, 1999). The ericoid mycorrhizas of dwarf shrub roots have been suggested to accumulate heavy metals, leading to a reduction in heavy metal concentrations in the shoot (Bradley et al.. 1981, 1982; Burt. 1984). Especially long-lived plants can avoid metals by proliferating roots in uncontaminated zones of the soils (Turner and Dickinson. 1993). However, even if the rate of heavy metal uptake can be controlled by plants, total avoidance of metal uptake is not possible (e.g. Baker. 1981). Tolerance to heavy metals can either be based on the evolution of tolerant genotypes, it can occur without evolution or it may be environmentally induced in the adaptation of plant populations to toxic soils (Antonovics et al., 1971; Baker. 1987; Wu, 1990). Bradshaw and Hardwick (1989) have defined constitutive adaptation (classical evolutionary change) as an inevitable evolutionary change in the population in a situation where particular stress is oc curring consistently, providing that an appropriate genetic variability exists in the popula tions affected. Facultative adaptation (phenotypic plasticity), in contrast, is produced within single genotype. The ways to achieve the adaptation in constitutive and facultative systems are fundamentally different, but there are not necessarily any differences in the physiological mechanisms (Bradshaw and Hardwick, 1989). 3 Different plant species have different degrees of metal tolerance, and they may possess varying levels of genetic variability and innate plasticity for tolerance. These mechanisms may interact with different types of polluted environments determining the success or failure of plant colonization (Wu, 1990). Multiple resistance to several metals is usually associated with the co-occurrence of high levels of these metals in the soil (Gregory and Bradshaw, 1965). Figure 1. Resistance mechanisms of plants (modified after Tomsett and Thurman, 1988; Marschner, 1995). 1.2. Toxicity of Cu and Ni and parameters used to evaluate the responses of plants to heavy metals 1.2.1. Toxic Cu and Ni concentrations for plants Although copper is an essential element for plants and participates in many metabolic proc esses of the cell (Marschner, 1995), elevated concentrations are toxic. Allaway (1968) re ported that a Cu concentration in leaf tissue of above 20 ng g" 1 dw is generally toxic to ter restrial plants. For most crop species a concentration of 20 ng g* 1 dw is toxic (Beckett and Davis, 1977; Davis and Beckett, 1978), whereas the yield was zero in the concentrations of 4 40-100 ng g" 1 Cu d\v (Beckett and Davis. 1977). Balsberg Pählsson (1989) defined a Cu concentration in the leaves of 15-20 \ig g" 1 dw as being toxic for plants. However, the toler ance varies widely between species, for example sitka spruce (Picea sitchensis) is compara tively insensitive, tolerating 88 fig g" 1 Cu dw in the needles (Burton and Morgan. 1983) Ni is also a mineral nutrient for higher plants, and so far it is known to be a component of one enzyme (urease) (Marschner. 1995). Leaf concentrations of 55 jig g'1 dw have been re ported to be potentially toxic for plants (Allawav. 1968). The critical concentration varies between species, for example, the upper critical Ni concentration for sitka spruce has been determined to be 6 ng g" 1 dw (Burton and Morgan, 1983), and for crops around 11-16 jig g": Ni dw (Beckett and Davis. 1977; Davis and Beckett. 1978). The lethal concentration is much higher; the zero yield of barley was obtained in the tissue concentration of 250-850 \xg g" 1 Ni dw (Beckett and Davis. 1977). The tissue of hvperaccumulators growing mainly in serpentine soils may contain 10 000 - 30 000 \ig g" 1 dw (Lee et al., 1978). 1.2.2. Elongation and biomass According to Baker and Walker (1989), the tolerant individuals can be separated from the non-tolerant individuals in their ability to establish, survive and reproduce in metal contaminated substrates. Growth in terms of biomass or length growth provides information about intraspecific differences in the effect of metal treatments. Root growth is usually rather sensitive to the presence of metal toxins (Baker and Walker. 1989). and the root elon gation test has been used to determine the metal tolerance index of higher plants (Wilkins. 1978). The root elongation test is based on the assumption that tolerance will be manifested as a genotype/environment interaction. When the concentration of the metal in the solution increases, the root growth of tolerant plants will be less affected than that of the less tolerant plants. The commonly used technique is the Tolerance Index (TI). which is root growth in a toxic solution / root growth in a control solution (Macnair. 1990). 1.2.3. Physiological measurements In addition to morphological measurements, several physiological parameters such as the chlorophyll, abscisic acid (a plant hormone), organic acid contents, water potential, photo synthesis of plants etc. have been used to indicate the responses of plants to elevated heavy 5 metal levels. Photosynthetic rate and chlorophyll concentrations of plants has been found to be decreased by Cu, Ni. Cd and Pb (Lamoreaux and Chaney, 1978; Becerril et al., 1989; Angelov et al.. 1993; Bishnoi et al.. 1993; Pandolfini et al., 1996). and connections between drought stress and Ni, Zn and Cd have been reported (Rauser and Dumbroff, 1981; Bishnoi et al., 1993). In experimental studies, the stomatal conductance and water potential of leaves have decreased and the ABA content, which regulates the water status of plants, increased when plants were exposed to Ni (Rauser and Dumbroff. 1981; Bishnoi et al.. 1993). Organic acids play a central role in detoxifying metals in Ni- and Zn-accumulating plants (Ernst, 1975; Mathys. 1977; Lee et al.. 1978; Yang et al., 1997), and the amount of organic acids in plant parts usually increases with increasing Zn or Ni concentrations in the soil (e.g. Lee et al.. 1978). 1.2.4. Electron microscopical investigations The ultrastructural localization of a large number of nutrients and heavy metals in plants is nowadays possible by electron dispersive X-ray spectroscopy (EDXS) and electron energy loss spectroscopy (EELS) (e.g. Mullins et al.. 1985; Turnau et al.. 1993 a. b; Neumann et al., 1995. 1997; Lichtenberger and Neumann, 1997). although both methods have their ad vantages and limitations (Stelzer and Lehmann, 1993; Kottke, 1994; Bucking et al., 1998). Very accurate information is obtained about the heavy metal resistance of plants when the metals are localized in specific cell compartments. Connections between the location of metals and nutrients in cell organelles or tissues provide information about detoxifying sub stances. However, the plant under study usually has to contain relatively high amounts of metals for them to be detected. 1.3. Ecology and special characteristics of dwarf shrubs colonising heavy metal contaminated areas 1.3.1. Ecology of Empetrum nigrum Crowberry (Empetrum nigrum (L.) ssp. nigrum and ssp. hermophroditum) is a xeromorphic (Miller. 1975; Carlquist, 1989; Wollenweber et al. 1992). evergreen dwarf shrub with an ericoid mycorrhiza (Read, 1983). It is characterised by clonal growth. It is wind pollinated 6 and black berries are long distance dispersed by animals and birds passing through their gut (Bell and Tallis. 1973). E. nigrum occurs on a variety of substrates. It can tolerate soil pH values ranging from 2.5 to at least 7.7 (Bell and Tallis, 1973). In the northern hemisphere E. nigrum is limited to the cooler regions, but it is also common in mountains at lower latitudes (Good. 1927; Bell and Tallis. 1973). In the subarctic plant communities it is often dominant species (Ojala. 1991). In boreal forests it grows on nutrient-poor, light dry heaths (Sarvas. 1937). and can form single-species plant communities in clearcut areas (Nilsson, 1992). It also grows on ombro genous peat and peaty podsols (Good. 1927: Bell and Tallis. 1973; Eurola et al.. 1990; Laine and Vasander. 1990) and serpentine soils (Proctor and Woodell, 1971). The ecology of E. nigrum has been widely studied in both boreal forests and arctic ecosys tems (Good. 1927; Bell and Tallis, 1973; Maimer and Nihlgard. 1980; Elvebakk and Spjel kavik, 1995; Michelsen et al., 1996 a, b; Tybirk et al., 2000). One reason for this is its spe cial ecology (dominance) in boreal forest ecosystems (Nilsson et al., 1998). One interesting feature of E. nigrum is its ability to influence the establishment of other species (Zackrisson et al.. 1997); its leaf extracts inhibit, for example, the establishment of Scots pine (Pinus sylvestris L.) seedlings (e.g. Nilsson and Zackrisson. 1992; Zackrisson and Nilsson, 1992; Nilsson, 1994). However, this feature is not typical for Empetrum species alone, because al lelopathic interactions between other plant species and trees have been also found in conif erous forests (e.g. Pellissier and Souto. 1999). Tybirk et al„ (2000) summarizes profoundly the function and susceptibility of E. nigrum dominated ecosystems to environmental changes. E. nigrum is relatively tolerant to simu lated acid rain (Shevtsova. 1998), tropospheric ozone (Johnsen et al.. 1991) and it responds relatively slowly to enhanced UV-B radiation (Johanson et al.. 1995). It is a weak competi tor for light (Tybirk et al.. 2000) and sensitive to waterlogging (Bell. 1969). mechanical disturbances and fire (Tybirk et al., 2000). Gerdol et al., (2000 a) found that length growth of the current-year shoots does not vary considerably between years though interannual fluc tuations occur. Shevtsova (1998) reported also, that the number of interactions between the effects of warming and watering have an effect on the growth and reproduction of E. nig rum. 7 1.3.2. Ecology of Calluna vulgaris The heather (Calluna vulgaris (L.) Hull.) dominated oceanic heathland ecosystems are per haps the most widely studied among dwarf shrub dominated ecosystems (e.g. Gimingham, 1972). C. vulgaris is a widespread and common species in Europe that usually grows on acidic and nutrient-poor soils (Gimingham. 1960, 1972). It is found on heathlands (e.g. Mohamed and Gimingham, 1970; Haapasaari, 1988). moors, bogs, fixed sand dunes and in forests (Gimingham. 1960). In Finland, C. vulgaris is one of the 10 most common vascular plant species in forest and peatland vegetation (National Forest Inventory 1995, unpublished results). C. vulgaris reproduces both vegetatively and from seeds (e.g. Gimingham, 1960, Mallik et al., 1988). Although C. vulgaris produces a vast number of seeds each year and there exists a huge seed reserve in the soil, the germination and subsequent establishment of these plants do not occur very readily (Mallik et al., 1984). Environmental factors and the seedbed sub strates affect seed germination (Mallik et al., 1988). C. vulgaris contains high amounts of carbon-based secondary metabolites such as phenolics and tannins (lason et al., 1993), and therefore the litter of C. vulgaris is rich in phenolic compounds (Jalal and Read. 1983). which decompose slowly and modify the soil causing acidification and organic matter accumulation. This leads to soil conditions that are optimal for the growth of C. vulgaris and impairs the growing conditions of other species (Grubb and Suter. 1971; Robinson, 1971; Haslam, 1977; Leake. 1988). The dominance of C. vulgaris in nutrient-poor environments has been explained on the ba sis of the effective use of nutrients; the ericoid mycorrhizal endophyte of C. vulgaris roots is also able to utilise organic sources of nitrogen (Bajwa et al., 1985; Leake and Read. 1989). However, contrary to the most of the studies, no significant differences between mycorrhizal and non-mycorrhizal plants in N concentration of C. vulgaris shoots were observed in ex perimental studies (Strandberg and Johansson, 1999). Long-term studies and greenhouse experiments have shown that an increased nitrogen load first strongly increases the shoot biomass of C. vulgaris (Carroll et al., 1999; Hartley and 8 Amos. 1999), but might be harmful in the long run by increasing the susceptibility of C. vulgaris to other environmental stresses such as frost and drought (Carroll et al.. 1999). 1.3.3. Dwarf shrubs in heavy metal polluted areas Species of the family Ericaceae ja Empetraceae colonise soils with naturally elevated metal concentrations such as serpentine soils (Proctor and Woodell, 1971; Marrs and Bannister. 1978) and soils contaminated with metals (Laaksovirta and Silvola. 1975; Marrs and Ban nister. 1978; Bagatto and Shorthouse. 1991; Bagatto et al.. 1993; Chertov et al., 1993; Lukina et al., 1993; Helmisaari et al.. 1995; Shevtsova. 1998: Mälkönen et al., 1999; Rei mann et al., 1999; Salemaa et al.. 2000 a, b; Uhlig et al., 2000) as a result of mining and smelting activities. However, also very resistant ecotypes and metal hyperaccumulators are found among herbs and grasses (e.g. Amiro and Courtin, 1981; Boyd et al.. 1999). and for example near the Sudbury Cu-Ni smelter in Canada, the grasses, herbs and decidious trees are the dominant vegetation in the polluted soils (Freedman and Hutchinson. 1980; Winter halder 1995; Bagatto and Shorthouse, 1999). The situation for trees and other long-lived perennial and clonal plants is different from many annual or colonizing plants, because the long generation time prevents rapid selection for tolerance (Dickinson et al.. 1991). Among dwarf shrubs. Vaccinium angustifolium is one of the plants surviving near the Cu-Ni smelter at Sudbury, Canada (Bagatto and Shorthouse, 1991; Bagatto et al., 1993; Shorthouse and Bagatto. 1995), and Vaccinium uliginosum and Arctostaphylos uva-ursi near the Cu-Ni smelter at Harjavalta, SW Finland (Salemaa et al., 1999). E. nigrum occurs in the immedi ate vicinity of the Cu-Ni smelter at Harjavalta (Helmisaari et al., 1995; Uhlig et al., 2000; Mälkönen et al., 1999; Salemaa et al.. 2000 a. b) and those in the Kola Peninsula. Russia (Chertov et al.. 1993; Lukina et al.. 1993; Shevtsova, 1998; Reimann et al., 1999). In most of the publications it is described as a species resistant to metals (Väisänen, 1986; Helmisaari et al., 1995; Shevtsova, 1998; Salemaa et al., 2000 a. b). C. vulgaris, however, is reported to be resistant or sensitive depending on the geographical region (Gilbert. 1975; Marrs and Bannister. 1978; Eltrop et al., 1991; Salemaa et al.. 2000 a. b). In Great Britain and Germany, C. vulgaris has been found to be a resistant species occurring on a variety of substrates, e.g. serpentine and polluted soils (Marrs and Bannister. 1978; Eltrop et al.. 1991) and colonising waste heaps in metal-polluted areas (Burt. 1984). In the northern parts of its 9 distribution C. vulgaris is absent from severely heavy metal polluted areas in Fennoscandia (Salemaa et al.. 2000 a. b) and the Kola Peninsula (Mikhail Kozlov, unpublished results), and it is considered to be a species sensitive to metals (Laaksovirta and Silvola. 1975. Sale maa et al.. 2000 a. b). The soils that are colonised by the dwarf shrub species are generally acidic and nutrient de ficient (Bell and Tallis, 1973: Marrs and Bannister. 1978). It is thought that dwarf shrubs are generally adapted to grow in soils where metal availability is high, and they therefore must have an intrinsic ability to tolerate high levels of metals (Meharg and Cairney, 2000). The resistance of C. vulgaris and Vaccinium species to metals has been suggested to be based on the ericoid mycorrhiza of the roots (Bradley et al., 1981. 1982; Burt, 1984). 10 1.4. Aims The overall aim of this thesis is to provide more information about the heavy metal resis tance of two important, common dwarf shrubs. E. nigrum and C. vulgaris, in boreal forests. These species were chosen because the former grows near smelters, whereas the latter is ab sent from the polluted areas mentioned in the above. The sensitivity and tolerance to metals were studied using greenhouse experiments and field studies. The greenhouse studies were carried out in order to eliminate factors other than specific heavy metals (I. 11, III), and the field studies to determine the actual situation in nature (IV, V). The specific aims of the thesis were; • to study the chemical composition, growth, biomass and discoloration of leaves of dwarf shrubs in response to elevated concentrations of Cu and/or Ni applied to the roots of E. nigrum (I) and C. vulgaris (II) in greenhouse conditions. • to study the chemical composition and aboveground element uptake of E. nigrum after nutrient and heavy metal applications to the aboveground parts. Ecophysiological re sponses (chlorophyll concentration, water potential, ABA content, dark respiration and maximum photosynthesis) to the applications were also studied (III). • to determine the responses of E. nigrum to heavy metal pollution by measuring physio logical parameters (chlorophyll contents, organic acids, stem water potential and ABA contents of E. nigrum leaves and stems) in the high and low contaminated areas near the Cu-Ni smelter at Harjavalta (IV). • to study the localization of heavy metals and nutrients in E. nigrum growing in the high and low contaminated areas near the Cu-Ni smelter at Harjavalta (V). 11 2. MATERIALS AND METHODS The materials, measured parameters and implementation of studies in papers I-V are shown in Table 1. Table 1. The materials, measured parameters and implementation of studies in papers I-V. 'Only Cu and Ni concentrations were increased in the treatment solution series, the nutrient concentrations being kept constant. 2.1. Greenhouse experiments The greenhouse experiments were carried out in the greenhouse of the Finnish Forest Re search Institute at Ruotsinkylä (60°21' N, 25°00' E). The light conditions were natural, and the mean temperature was regulated at +2O-22 °C during the day and +l5 °C at night. The mean relative humidity in the greenhouse was approximately 60 to 70% (I. 11, III). The three- to five-year-old (I) and four- to five-year-old (III) E. nigrum cuttings originated from an unpolluted area in SW Finland. The cuttings were rooted in a mixture of sand and peat and transferred to quartz sand for the Cu and Ni experiment (I) or grown in quartz sand Paper Species Parameters Elements applied Implementation 1 E. nigrum Chemical composition, elonga- tion, biomass *Cu, Ni, Cu+Ni + nutrients Greenhouse II C. vulgaris Chemical composition, elonga- tion, biomass *Cu + nutrients Greenhouse III E. nigrum Chemical composition, chloro- phyll, C02-exchange rate, stem water potential, ABA Cu, Ni, Pb, Zn, Fe, Cd, Cr, Mn, Mg, Ca, P, K, C, N Greenhouse IV E. nigrum Chlorophyll, organic acids, stem water potential, ABA, soil moisture Field; 0.5 and 8 km dis- tances from the smelter V E. nigrum Electron microscopical element localization Field; 0.5 and 8 km dis- tances from the smelter 12 (on the bottom) and peat substrate for the aerial heavy metal and nutrient application (Cu, Ni, Fe. Pb. Zn. Cd. Cr. Mn. Mg, Ca. P. K, N) experiment (III). In Cu and Ni experiment (I) the cuttings were watered twice a week (50 ml/pot) with a nu trient solution modified by Stribley and Read (1976) containing P. K. Ca. Mg, Fe. Mn, B. Zn, Mo and NH4- and NO3-N. Six levels of Cu (0.1, 1. 10, 22. 46 and 100 mg l" 1 ). five levels of Ni (0, 10, 22. 46 and 100 mg l" 1 ) and nine combinations of Cu and Ni [Cu (mg 1"')/Ni (mg I -1 ): 10/10. 22/10. 46/10. 100/10. 22/22. 46/22. 100/22. 46/46. 100/46] were used. Cu was given as CIISO4 and Ni as NiCk The experimental design was completely random, and the total number of seedlings treated was 152 (8 plants/treatment) (I). In the aerial heavy metal and nutrient application experiment (III) solutions with six differ ent compositions of 13 elements were sprayed twice a week on the seedlings. The aerial parts of the seedlings were enclosed in a plastic container, the spraying treatment being applied via openings in the container. The seedlings were sprayed from two opposite direc tions with 15 ml of solution from each direction (altogether 15 times). The composition of wet deposition near the Cu-Ni smelter at Harjavalta, SW Finland, was simulated. Treatment I represented control, the nutrient concentrations being the same in treatments I and 11. Treatment II contained about the same concentration of nutrients and heavy metals as rain water at a distance of 0.5 km from the smelter reported by Helmisaari et ai. (1994) and Helmisaari (personal communication). In treatments 111 to VI, the concentrations of all the components increased exponentially, the relative concentrations thus being constant in the different treatments (see table 1 in paper III). The experimental setup was completely ran dom and there were nine plants per treatment (III). The seed bank samples of C. vulgaris (II) were collected from three peatland sites located 1.2 km to the NW (Lammaistensuo) and 2.5 km (Kotosuo) and 5.5 km to the NE (Pyhäsuo) of the Harjavalta Cu-Ni smelter (61°19'N. 22°9'E). The seeds were germinated in a mixture of sand and peat and the nine-month-old seedlings planted in quartz sand for the Cu ex periment. The cuttings were watered twice weekly (50 ml/pot) and the same nutrients as above (I) were given (see plant culture system in paper II). Five concentrations of Cu (1. 10. 22. 46 and 100 mg l" 1 ). applied as CIISO4. were used with ten replicates per treatment. The 13 total number of seedlings was 150 (3 origins x 5 treatments) and the experimental setup was completely random (II). 2.2. The study sites near Harjavalta Cu-Ni smelter The Harjavalta Cu-Ni smelter, situating SW Finland (61° 19' N, 22°9' E), started operating 55 years ago. Cu smelter was established in 1945 and the Ni smelter in 1960. Sulphur diox ide and heavy metals have been emitted into the environment for the past 40-60 years. The deposition of metals near the smelter was considerably reduced in the 1990' s after a new taller stack and electrostatic filters were built (Rantalahti, 1995). The prevailing wind direc tion has been from the south, south-west and south-east (Derome, 2000). A detailed de scription of the emissions and smelter activity has been presented by Derome (2000). Samples of E. nigrum for the ecophysiological (IV) and electronmicroscopical (V) studies were collected at two distances (0.5 and 8 km) from the Harjavalta Cu-Ni smelter (Figure 2). For ecophysiological studies ten separate E. nigrum patches were marked situating more than four meters from each other, and these plants were used for all physiological measure ments at both locations (0.5 and 8 km). The material collection was done on July 21 st , 26 th , August 18 dl -19 ai . 26 th and October 9 th . 1997 (IV). The sampling for electronmicroscopical studies was done on August 6 , 1998 (V). The sites were chosen SE from the smelter, because it was the only direction where the two sites represented the same forest site type (Colluna site type) and soil type (orthic podzol) (Figure 2). The pH of the organic layer was 3.5 at 0.5 km distance and 3.6 at 8 km distance from the smelter (Derome and Lindroos, 1998 a). The sulphur deposition has not had an ef fect on soil acidity in the organic layer or upper mineral soil layers, although an increase in exchangeable acidity and A 1 in deeper mineral soil at 0.5 km was found (Derome. 2000). At 0.5 km distance the site is located in a heavily polluted area where the total Cu and Ni con centrations in the organic layer are over 5 800 and 460 mg kg' 1 dw respectively, which has resulted a displacement of Ca. Mg. K from exchange sites and a decrease in plant available concentrations of these cations. This is also caused by a partial inhibition of the mineralisa tion of these nutrients from the litterfall (Derome, 2000). The concentrations of other heavy metals (Fe. Zn, Cd. Pb. Cr) in the organic layer are also elevated near the smelter (Derome 14 Figure 2. Sampling sites at a) 8 and b) 0.5 km distance from the Cu-Ni smelter at Harjavalta, SW Finland. 15 and Lindroos, 1998 a; see also metal concentrations in article V). The site at 8 km is only slightly polluted, the total Cu and Ni concentrations in the organic layer being 150 and 40 mg kg" 1 dw. respectively. The concentrations of other heavy metals are also much lower at 8 km than at 0.5 km. but are higher than background values (Derome and Lindroos. 1998 a). There are several examples about the influence of heavy metal and sulphuric acid deposition as well as nutrient deficiency on the ecosystem (e.g. Helander, 1995; Fritze et al., 1996; Pennanen et al.. 1996; Nieminen et al., 1999; Mälkönen et al. 1999; Salemaa et al., 1999, 2000 b; Uhlig et al., 2000). For example Scots pine (Pinus sylvestris) suffers from Mg defi ciency (Nieminen et al.. 1999) near the smelter and the number of pine needles infected by specific endophyte, Cenangium ferrucinosum Fr.:Fr. is decreased (Helander, 1995). Also fine root mass (Mälkönen et al., 1999) and soil microbial and fungal biomasses of the soil are decreased (Fritze et al., 1996; Pennanen et al., 1996). At the site 0.5 km the vegetation is almost totally absent apart from the few patches of mosses (Pohlia nutans, Ceratodon pur pureus) and the dwarf shrub E. nigrum ssp. nigrum. Carex globularis and Vaccinium uligi nosum have also survived in partially paludified depressions. Abundances of many species at 8 km are rather typical for dry heath forests (Salemaa et al.. 2000b). 2.3. Chemical analyses The total concentrations of Ca, Cd, Cu, Fe, K, Mg, Mn, Ni, P, Pb and Zn in the peat (III) and different plant parts [stems, leaves by year growth (I, 11, III) and roots (II), detailed de scription see papers I, 11, lII] were determined by dry digestion (+550 °C), followed by ex traction of the ash with 2-3 ml of 6 M HCI (pro analysi). The solutions were analysed by in duction coupled plasma atomic emission spectrometry (ICP-AES) (I. 11. III). The C and N concentrations were determined on the dry material using a Leco CHN analyser (Nelson and Sommers, 1982) (III). The results are given as concentrations (mg kg" 1 ) (Timmer, 1991). 2.4. Physiological measurements For the analysis of chlorophyll a and b, freeze-dried plant material (Hetosicc freeze dryer, type CD 52) was extracted with 80% acetone and the absorbances (A) at 647 and 664 nm recorded on a spectrophotometer (Shimadzu UV-1201. UV-VIS) (Graan and Ort. 1984) (111, 16 IV). The results are calculated as chlorophyll concentration (|imol l"1 ) (III) or content in the tissue (nmol chlorophyll g" 1 dw) (IV). Citric and malic acids in freeze-dried material were determined enzymaticallv by a method modified from Boehringer (1989) and Hampp et al. (1984). The organic acids were ex tracted by 0.1 N HCI. and measured by a spectrophotometer at 340 nm (Kontron). The re sults are shown as contents (nmol mg" 1 dw) (IV). Pressure chamber determinations were carried out to estimate the total water potential of the xvlem sap of E. nigrum. The stems were cut with a razor blade, and the stem water potential measured using a Scholander pressure bomb (Scholander et al.. 1965; Ritchie and Hinckley, 1975; Richter. 1997) (111. IV). The soil moisture measurements were made at the same time as the stem water potential measurements. The soil moisture was measured using a ThetaProbe soil moisture sensor (type MLI). The ground vegetation was removed and the measurements were made horizon tally in three soil horizons (organic layer = O, mineral soil = A and B horizons) (IV). For the analysis of abscisic acid, freeze-dried (Hetosicc freeze dryer, type CD 52), homoge nised plant material was suspended in 0.05 M phosphate buffer (pH 8.0). and 50 ng internal standard (| 2 H. IjABA) was added. The residue was dissolved in 80% methanol and injected into a Radial Pak™ (Waters. Milford. USA) Cig HPLC column. The fraction containing ABA was collected (15.5-17.5 min), and reduced to dryness in vacuo. The residue was dis solved in methanol and methylated. The dried sample was dissolved in 12 jj.l of heptane, and 1 (.il of this solution was injected into the GC-MS. lons of m/z 162. 190. 193 and 194 were monitored. The results are shown as contents ng g" 1 (111. IV). Note that here the term con tent is used for (ng g" 1 ) unlike in chemical analyses the term concentration (mg kg" 1 ) is used. The CO2 exchange rate of E. nigrum was measured using a batten -operated Li-Cor LI-6200 portable photosynthesis system (Li-Cor Inc., Lincoln, NE, USA). Measurements were car ried out with a quarter litre chamber. Six different irradiance levels (0. 50. 110. 330. 600 17 and 820 fxmol m" 2 s" 1 ) of photosvnthetically active radiation (PAR) were used. Dark respira tion was measured by covering the chamber with a black plastic sheet (III). 2.5. Electron microscopy Previous-year stems of E. nigrum were collected in the field, cryofixed and stored in liquid nitrogen (-192 °C). The stem pieces were freeze-dried (CFD. Leica, Germany) and pressure infiltrated directly in 100 % Spurr's epoxy resin (Spurr. 1969), using a method described by Fritz (1980). For energy dispersive X-ray spectrometry (EDXS) the embedded samples were dry sectioned (0.5 (mi), placed on filmed Ni or Cu grids and carbon coated (V). The EDXS studies were carried out under standardised conditions using a Philips EM 420 provided with the ED AX DX-4 system. EDXS spectra were collected between 0 and 20 keV with a Si(Li) X-ray detector. EDXS point measurements were made in the xylem, phloem and parenchyma of the primary ray of the stems in both high and low contaminated samples (V). 2.6. Statistical analysis Two-factor ANOVA (analysis of variance) was used in analysing the effects of Cu and/or Ni or origin on the response variables of E. nigrum and C. vulgaris (GLM procedure. SAS In stitute Inc.. 1994; Sokal and Rohlf. 1995). Pairwise comparisons between the treatments and between the Cu and Ni series were performed by the t-test (I, II). Relationships between Cu and Ni concentrations in plant parts and the applied concentrations were studied by regres sion models (REG and NLIN procedures, SAS Inst.). Regression equations are also given for Cu and Ni uptake as a function of the amounts of Cu and Ni applied (SAS Institute Inc.. 1994; Sokal and Rohlf, 1995) (I). Pearson correlations were calculated between the concen trations of Cu and other elements in the different plant parts (II). Non-parametric Kruskal-Wallis ANOVA (analysis of variance) was used in analysing the effects of spraying treatments on the stem water potential of E. nigrum. Pairwise compari sons between the treatments were performed by the Kruskal-Wallis, comparison of mean ranks test. The Spearman rank correlations were calculated between Cu and Fe concentra 18 tions in different plant parts and stem water potential, chlorophyll and ABA contents of E. nigrum leaves. In order to determine the element uptake via the roots from the peat and possible contamination of the peat by the spraying solutions, the Spearman rank correlations between the element concentrations in peat and different parts of E. nigrum were calculated (Sokal and Rohlf, 1995; Statistix. 1996). The effect of treatment on the CO2 exchange rate of E. nigrum was evaluated by comparing the dark respiration at the irradiance level of 0 (imol m" 2 s" 1 and the maximum photosynthesis at the irradiance levels of 600 or 820 jrmol m" 2 s" 1 in the different treatments. One-way analysis of variance was performed and the dif ferences between the treatment means were compared by the t-test (SAS Institute Inc., 1994) (HI). In order to evaluate the effects of heavy metal deposition on E. nigrum derived from the two sites, the means of the measured parameters (soil moisture, chlorophyll, organic acid. ABA contents, stem water potential) (IV) and the means of different elements in cell compart ments (V) were compared by the t-test after logarithmic transformations had been performed to normalise the data. Otherwise the Kruskal-Wallis, comparison of mean ranks test, was used (Sokal and Rohlf, 1995; Statistix. 1996) (IV, V). 3. RESULTS 3.1. Chemical composition of E. nigrum and C. vulgaris in response to element applications (I, 11, III) Table 2 summarizes the maximum concentrations of heavy metals in different parts of living E. nigrum and C. vulgaris when exposed to heavy metals in greenhouse (I. 11, III). The maximum metal concentrations in plants were usually the result of the highest applied metal concentrations in the nutrient solution with some exceptions. The experimental setup and thus the maximum concentrations in nutrient solutions applied to the plant roots (I, II) or aboveground parts (III) see papers I-111. 19 Table 2. The maximum element concentrations in different parts of living E. nigrum and C. vulgaris after element applications to the roots (I, II) or aboveground parts (III). Species Metal applied Plant part Maximum concentrations in the livinq tissue mq kq " 1 C. vulgaris Cu to roots (II) old roots 2240 stems of the leading shoots 1370 stems of the side branches 710 discoloured leaves 710 new roots 540 green leaves 60 E. nigrum Cu to roots (1) older stems 2030 older discoloured leaves 470 younger stems 180 younger discoloured leaves 140 younger green leaves 80 Ni to roots (1) older stems 2130 younger stems 580 younger discoloured leaves 340 older discoloured leaves 310 younger green leaves 100 E. nigrum Cu to aboveground bark (older) 2150 parts (III) leaves (older) 260 stems (older) 80 Ni to aboveground bark (older) 160 parts (III) leaves (older) 20 stems (previous-year) 30 Fe to aboveground bark (older) 660 parts (III) leaves (previous-year) 250 stems (previous-year) 200 Pb to aboveground bark (older) 50 parts (III) leaves (older) 10 stems (current-year) 6 Zn to aboveground bark (older) 330 parts (III) leaves (older) 80 stems (current-year) 30 Cd to aboveground bark (older) 6 parts (III) leaves (older) 0.6 stems (older) 0.4 20 3.1.1. Accumulation of Cu and/or Ni in E. nigrum (I) During the six-week course of the experiment, E. nigrum accumulated increasing amounts of Cu and/or Ni with increasing Cu and/or Ni levels in the nutrient solution. Metal accumu lation increased with age, the younger parts containing less Cu and Ni than the older ones. The accumulation pattern was similar independant whether the Cu and Ni were applied separately or in combination, and it was similar over the whole concentration range. The highest Cu and Ni concentrations were measured in the oldest stems (Table 2). The roots were not analysed due to technical problems. 3.1.2. Accumulation of Cu, Ni, Pb, Zn, Fe, Cd, Mn, Mg, Ca, P, K, N in E. nigrum (111) The concentrations of Cu, Ni, Pb, Zn. Fe and Cd in the leaves and bark of E. nigrum gen erally increased due to increasing concentrations of those elements in the spraying solution. In contrast, the Mn. Mg, Ca, P. K and N concentrations did not increase in any plant parts in response to increasing nutrient applications. Cu, Ni, Pb, Zn, Fe and Cd accumulation was the highest in the older bark (Table 2) and increased with increasing age in the leaves and bark. In the stems, the age-dependant accumulation was not found (III). 3.1.3. Accumulation of Cu in C. vulgaris (II) During the six-week course of the experiment, Cu concentrations increased in all parts of C. vulgaris with increasing Cu levels in the nutrient solution. The highest Cu concentrations were measured in the old roots and stems of the leading shoots (Table 2). In the living plants the concentrations were considerably lower than in the dead plants (II). 3.2. Ecophysiological responses of E. nigrum to metals in the greenhouse and field (III, IV) 3.2.1. Photosynthesis of E. nigrum (111, IV) The sum of chlorophyll (a+b) did not change, while the chl alb ratio increased slightly with increasing element (Cu. Ni. Pb. Zn. Fe, Cd. Cr. Mn. Mg. Ca. P. K. N) applications in the 21 greenhouse (III). Unlike in the greenhouse, the plant chlorophyll (a+b) contents were lower at 0.5 than at 8 km from the Cu-Ni smelter, and the differences between the means were statistically significant throughout the whole season (p < 0.05) (IV). The means of the chlo rophyll (a+b) contents varied between 1.9 to 3.2 pmol chlorophyll g" 1 dw. and were the low est in October compared to the other sampling dates. The chlorophyll alb ratio was gener ally lower at a distance of 8 km than at 0.5 km. and the difference between the means at the two distances was statistically significant (p < 0.05) in mid August (IV). In the greenhouse the highest CO2 exchange rate of E. nigrum was measured at the irradi ance levels of 600 and 820 jimol m" 2 s" 1 . Increasing element concentrations (Cu. Ni. Pb. Zn. Fe. Cd, Cr. Mn, Mg. Ca, P. K. N) in the spraying solution decreased dark respiration and maximum photosynthesis of the E. nigrum current-year shoots (III). 3.2.2. Stem water potential and ABA of E. nigrum (111, IV) The increasing treatment levels in the greenhouse had no consistent effects on the water potential of E. nigrum, which varied between -11 to -13 bars (III). In the field the stem water potential of E. nigrum varied more and was more negative during the day (-15 to -21 bars) than at night (-4 to -12 bars) (IV). During the day, the water potential of plants at 8 km was more negative than those at 0.5 km. However, during the night the opposite occurred (more negative at 0.5 km). The means of the stem water potential measured during the day and at night in July and during the day in August differed significantly between the plants growing at the two sites (p < 0.05) (IV). In the greenhouse the abscisic acid contents varied between 13 to 66 ng g" 1 dw. The spraying treatment appeared to affect the ABA contents, the highest contents being in treatments IV and VI, and the lowest values in treatments II and 111 (III). In the field, the abscisic acid content varied more (IV) than in the greenhouse studies (III). The mean ABA contents in the leaves varied between 66-232 ng g" 1 dw and 56-196 ng g" 1 dw. and in the stems between 45-113 ng g" 1 dw and 33-71 ng g" 1 dw at 0.5 and 8 km. respectively. Empetrum plants growing at 0.5 km had higher abscisic acid contents in their stems compared to those growing at 8 km. With the exception of the July sampling, a similar pattern was also ob 22 served in the leaves. No statistical differences were found between the means (p > 0.05). In late autumn the ABA contents decreased in the stems, but not in the leaves (IV). 3.2.3. Organic acid contents of E. nigrum in the field (IV) The citric acid contents in the leaves and stems of E. nigrum were higher at 8 km than at 0.5 km. In the stems, the difference between the means was statistically significant (p < 0.05). In the leaves, the mean pools of citric acid were 16 nmol mg" 1 dw at 8 km and 13 nmol mg" 1 dw at 0.5 km, and thus exceeded those in the stems of 14 nmol mg" 1 dw and 11 nmol mg" 1 dw, respectively (IV). The pools of malic acid in the leaves and stems of E. nigrum were higher at 8 km than at 0.5 km. The difference between the means of the malic acid content of the stems was statisti cally significant (p < 0.05). In the leaves, the mean malic acid values exceeded those in the stems, and were 17 nmol mg" 1 dw at 8 km and 14 nmol nig" 1 dw at 0.5 km. compared to 13 nmol mg" 1 dw and 8 nmol nig" 1 dw. respectively, in the stems (IV). 3.3. Ultrastructural element localization of E. nigrum in the field (V) The heavy metal localization in stems of E. nigrum analysed by EDXS varied according to the metal and tissue, and the amounts (peak to background ratio) of Cu. As and Fe were higher near to the smelter than in the stems from the control site. The highest Cu amounts were measured in the vessel lumens and primary wall of the ray cells, and Cu amounts were elevated both in living (ray cells, phloem, parenchyma of the primary ray) and in dead cells (xylem, sclereids). Cu was rather evenly distributed among the tissue. The highest As amounts were measured in the outer regions of the stem cross-section and lower As amounts were generally detected in vessels and tracheids of the xylem. In general, ray cells, phloem and sclereids had higher Fe amounts than other tissues from the contaminated stem samples. Unlike in the Fe amounts, the A 1 amount in many cell compartments was higher in the highly contaminated site, but the difference between the two distances was very small, and the variation could not be explained by the effect of the site. The highest A 1 amounts were detected in the electron-dense material of the ray cells. In contrast to the elements described above, the Mn amounts in the different cell compartments of the stems were lower near to the smelter. High Mn amounts were detected in low contaminated samples, especially in 23 living tissue (ray cells, phloem, parenchyma of the primary ray) and the cell walls of sclereids of the stem. In contrast, the amounts were very low in the xylem of the vascular tissue (V). The amounts of macronutrients varied in the different cell compartments. The Ca amounts were generally higher at 8 km than at 0.5 km distance from the smelter, the highest amounts occurring in the electron-dense material of vessel lumens. Ca was mainly located in the liv ing parts of the tissue, in ray cells, phloem and parenchyma of the primary ray. and espe cially in the primary cell walls. The amounts in the primary cell walls of dead tissue (vessels, tracheids, sclereids) were lower. The K amounts were higher at 0.5 km than at 8 km. The highest K amounts were measured mainly in the living parts of the tissue, i.e. in the phloem and parenchyma of the primary ray. The P and S amounts did not van,- accord ing to the site. There were relatively high P amounts in the ray cells, phloem and paren chyma of the primary ray, whereas only low amounts were detected in the vessels and tra cheids of the xylem and sclereids. In the living tissue of the stem cross-section. P and S were mainly located in the cytoplasm or electron-dense material. In contrast, only low amounts were found in the vacuoles and lumens (V). 3.4. Growth of E. nigrum and C. vulgaris in response to Cu and/or Ni appli cations (I, II) 3.4.1. The elongation and biomass of E. nigrum in response to Cu and/or Ni (I) Elevated Cu or Ni concentrations in the nutrient solution decreased the elongation of the shoots and the dry weights of the current-year shoots (leaves and stems combined) and roots, and increased the biomass of discoloured leaves of E. nigrum. The maximum growth reduc tion of the elongation of the leading shoot and side branches was 72-79% when treated with 100 mg l" 1 Cu. The corresponding reductions for 100 mg l" 1 Ni were 96% and 85%. Ni therefore reduced elongation more than Cu. When Ni was added to the highest Cu level (100 mg l" 1 ) there were no differences in length growth compared to the plants treated only with Cu (100 mg l" 1 ). The overall survival was not affected (I). 24 Both Cu and Ni suppressed biomass production. Compared to the highest dry weight, the dry weight of the current-year shoot was limited by 78 and 79 % in plants treated with 100 mg l" 1 Cu or Ni, respectively. At low Cu levels (Cu <46 mg l" 1 ) Ni increased the growth sup pression, but at higher Cu concentrations (46, 100 mg l" 1 ) there was no effect (I). Root growth was affected already at relatively low levels of Cu and Ni. The root growth of E. nigrum was only about 1 to 4% of the total belowground biomass when the Cu or Ni con centrations in the nutrient solution were more than 22 mg l" 1 . When a combination of the two metals was given, the maximum proportion of new root dry weight out of the total be low ground biomass was 6% (46 mg 1"' Cu and 10 mg l" 1 Ni applied) (I). 3.4.2. The elongation and biomass of C. vulgaris in response to Cu (II) In greenhouse experiment, the growth and survival of C. vulgaris were strongly decreased due to Cu application. The maximum growth reduction of the leading shoots of C. vulgaris was 86-99%. and that of the side branches 84-92%, compared to that for the control treat ment (1 mg l" 1 Cu) (II). The total shoot biomass of C. vulgaris decreased less than the length growth, the maximal reduction in biomass varying from 52 to 67% in the different seedling origins. The biomass of discoloured leaves of C. vulgaris increased with increasing metal applications (II). Root growth was affected already at relatively low levels of Cu. Root growth of C. vulgaris was almost totally inhibited by 100 mg l" 1 Cu, the proportion of new roots being only 2-3% at the highest Cu concentration (II). 25 4. DISCUSSION 4.1. Heavy metal accumulation in C. vulgaris and E. nigrum in controlled condi tions 4.1.1. Measured concentrations The greenhouse studies showed that maximum concentrations in the living roots (over 2000 mg kg" 1 Cu dw) and stems (over 1000 mg kg' 1 Cu dw) of C. vulgaris were at the same level as those reported in the experimental study of Burt (1984) on non-mycorrhizal plants. Bradley et al. (1981, 1982) reported even higher shoot and root concentrations in mycorrhi zal and non-mycorrhizal plants. The Cu concentrations in plants growing in Cu-polluted soil have been considerably lower. In Great Britain, the Cu concentration in C. vulgaris fine roots was only about 140 mg kg" 1 and in brown shoots 90 mg kg" 1 , even though the total Cu concentration in the spoil heap was 2500 mg kg" 1 (Burt. 1984). 115 mg kg" 1 Cu has been re ported in green shoots (Marrs and Bannister. 1978) (II). The Cu and Ni concentrations (over 2000 mg kg" 1 ) in the different plant parts of E. nigrum, when applied to the roots, were higher than those measured earlier in the living tissue of E. nigrum. The highest reported Cu concentration in older living parts has been 1450 mg kg" 1 (Helmisaari et ai., 1995) and the Ni concentration in leaves 1510 mg kg" 1 in the field (Chertov et al., 1993) (I). When Cu was applied to aboveground parts, the bark contained a maximum of over 2000 mg kg" 1 Cu even though the tissue was washed with distilled water before analysis (III). These measured concentrations are high compared to e.g. barley of which the zero yield was obtained in the tissue concentration of 250-850 jag g" 1 Ni dw or 40- 100 pg g" 1 Cu dw (Beckett and Davis, 1977). However, the tissue of hyperaccumulators growing mainly in serpentine soils may contain 10 000 - 30 000 \ig g" 1 Ni dw (Lee et al., 1978). Because the metal concentrations (Cu. Fe. Pb. Cd. Ni and/or Zn) increased with increasing treatment levels when metals were applied to the roots (I, II) or to the aboveground parts (III) of the dwarf shrubs, these results suggest that not only root uptake of Cu and Ni but 26 also surface contamination significantly contribute to the levels of these metals in the above ground parts of plants exposed to aerial pollution. 4.1.2. Accumulation pattern The Cu concentrations in C. vulgaris generally decreased in the order (also dead plants in cluded): old roots > new roots > stems > discoloured leaves > green leaves (II). When Cu and Ni were applied to the roots of E. nigrum, the accumulation pattern of Cu and Ni was stems > discoloured leaves > green leaves. Roots of E. nigrum were not studied but in the field relatively lower concentrations of Cu have been found in the roots compared to the older tissues (Salemaa and Vanha-Majamaa, 1998; Uhlig et al., 2000). The accumulation pattern of these metals within E.nigrum was in good agreement with the values reported for E. nigrum growing in the field (Uhlig et al., 2000) (I). When metals (Cu. Fe. Pb. Cd. Ni. Zn) were applied to the aboveground parts the concentrations decreased generally in the following order: bark > leaves > stems (III). The accumulation pattern in all the studies (I, 11, III) was similar over the whole concentration range. Because this pattern was not re stricted to the highest metal applications, it seems that root-to-shoot transport as well as re stricted transport to the green leaves occur (I, II). Increasing heavy metal concentrations with age is a pattern generally found in the above ground parts of E. nigrum growing in the field (Helmisaari et al., 1995; Uhlig et al., 2000). Both greenhouse studies confirmed the results obtained in the field, suggesting that E. nig rum accumulates metals in its older tissues especially (I. III). The accumulation pattern seems to be the same irrespective of whether Cu and Ni are applied to the roots (I) or to the aboveground parts (III). The stems of other dwarf shrub species. Vaccinium uliginosum and Arctostaphylos uva-ursi, have been found to have higher concentrations of Cu in the older than in the younger parts. However, the difference in concentrations between the age classes for A. uva-ursi is not as high (Salemaa and Vanha-Majamaa, 1998) as for E. nigrum. The strong surface binding of heavy metals in the older tissue of E. nigrum (III) is also in good agreement of the results of Uhlig et al. (2000), who found that the litter of E. nigrum con tains considerable amounts of heavy metals. 27 4.2. Ecophysiological responses of E. nigrum to elements in greenhouse and field conditions 4.2.1 Photosynthesis of E. nigrum The chlorophyll contents of E. nigrum were not affected by the aerial application of a mix ture of heavy metals (Cu, Fe. Pb, Cd. Cr, Ni, Zn) in the greenhouse (III). At 0.5 km from the smelter, where high heavy metal concentrations enter the plant from the toxic soil via the roots, the chlorophyll concentrations of E. nigrum were 8 to 30% lower than the values for plants growing at 8 km (IV). Similar decreases in chlorophyll concentrations have been re ported for several broadleaved species exposed to metals (Angelov et al., 1993; Bishnoi et al., 1993). The chlorophyll alb ratio increased slightly with increasing metal treatments in the greenhouse (III), and this was also seen in the E. nigrum leaves near to the Cu-Ni smelter in the field (IV). The chlorophyll alb ratio is used as a stress indicator and has been reported to increase as a result of environmental stress (Delfine et al., 1999). The light saturation for E. nigrum occurred between 600 and 800 p.mol m" 2 s" 1 and was about the same as found for Vaccinium angustifolium (500 - 600 uniol m" 2 s" 1 ) (Hicklenton et al., 2000). Although the chlorophyll concentrations were not affected in a study where a mixture of metals was applied to the aboveground parts (III), there were signs of a decrease in dark respiration and maximum photosynthesis due to the aerial application. No positive correlation was found between the applied N concentrations and CO2 exchange, although an increased N content generally increases the photosynthetic rate (Hoogesteger and Karlsson, 1992). The positive correlation between leaf-N concentration and of Vaccinium vitis-idaea and Vaccinium myrtillus has also been found (Gerdol et al., 2000b). However, despite the high N concentration in the spraying solution, the leaf N concentration of E. nig rum did not increase (III). Also the form of nitrogen applied affects; CC>2-exchage rate of Vaccinium corymbosum L. was higher when the roots were supplied to ammonium nitrogen with or without calciumcarbonate than in a case when nitrate nitrogen was applied (Claussen and Lenz, 1999). 28 4.2.2. Water stress of E. nigrum The mixture of heavy metals had no clear effect on the water potential of E. nigrum in the greenhouse (III). The fact that heavy metal pollution affected the water potential in the field (IV) indicates that the metals applied to the aboveground parts remained on the surface of E. nigrum (III). In contrast, the ABA contents of E. nigrum leaves and stems increased slightly with increasing metal treatments (III) or deposition in the field (IV). the results being partly in agreement with earlier studies showing heavy-metal induced ABA accumulation (Rauser and Dumroff. 1981; Poschenrieder et al., 1989). In manipulation studies the ABA content increased 2-fold as a result of excess Ni in bean leaves (Rauser and Dumbroff 1981). This increase is similar to that in the E. nigrum leaves exposed to the highest heavy metal treat ment (III). The overall ABA contents of the E. nigrum leaves were lower in the greenhouse experiment (III) than in the field (IV). The time of the year also affected the ABA contents when measured in the field. The ABA contents were high in the leaves in October, which may be related to environmental conditions, e.g. to decreasing temperature and ageing of the leaves (IV). The ABA content in the leaves usually rises as the leaf water potential falls (Beardsell and Cohen. 1975). The leaf ABA content was higher and stem water potential lower at 0.5 than at 8 km distance from the Harjavalta smelter in July, but not in August. However, in the early stages of soil desiccation, ABA is transported as a chemical signal from the roots to the leaves where, by enhancing stomatal closure, it prevents a decline in the water potential (Davies and Zhang, 1991). Furthermore, the production of ABA in nutrient-deficient plants seems to occur at less negative water potentials (Radin. 1984). Near to the smelter at Har javalta, E. nigrum is suffering from low concentrations of Mg and Mn. although the concen trations of other nutrients (N, P, Ca, K) in the plant tissue are not significantly lower (Uhlig et al., 2000). In addition to nutrient deficiency, the plants may require less water for growth and photosynthesis as these metabolic processes have declined. The roots, where ABA syn thesis occurs (Marschner, 1995), are damaged near to the smelter and this could, in turn, affect ABA synthesis and subsequently the measured ABA contents (IV). 29 4.2.3. Organic acid contents of E. nigrum The organic acid contents were measured in the field in order to determine whether there is any connection between organic acid and Ni and Zn concentrations in the plant. However, the organic acid contents in E. nigrum stems and leaves were lower in the plants growing at 0.5 km than at 8 km distance from the smelter, and therefore the results of our study do not follow the generally accepted pattern. Elevated pools of organic acids are normally associ ated with elevated heavy-metal levels in the substrate (especially Ni and Zn) (Lee et al., 1978; Godbold et al.. 1984; Yang et al., 1997). Organic acids are known to take part in the uptake and transport of metals, and to accumulate in the cytosol or vacuoles of plants (Ernst. 1975; Ernst et al., 1992). For this reason, their concentrations increase in plant parts as a results of Ni or Zn stress (Lee et al., 1978; Godbold et al.. 1984; Yang et al., 1997). At Har javalta. however, the lower metabolic efficiency measured by the decreased chlorophyll contents might explain the decreased production of organic acids near the smelter (IV). 4.3. Ultrastructural localization of elements in E. nigrum The higher Cu, As and Fe amounts in the cellular compartments of stems of E. nigrum col lected from the highly contaminated area compared to the low contaminated plots were in good agreement with greenhouse study (I) and earlier results obtained near to the Harjavalta Cu-Ni smelter (Helmisaari et al., 1995; Uhlig et al.. 2000). Although arsenic concentrations in the soil and plants have not been earlier reported, the emissions from the smelter do in fact also contain As (Rantalahti, 1995). The studies of Derome and Lindroos (1998 a) and Uhlig et al. (2000) support our results about higher Mn amounts in different cell compart ments at 8 km than at 0.5 km. The cellular K. P and S amounts by EDXS reflected the macronutrient concentrations in the plant parts at the two sites, the nutrient concentrations (Ca. K. P and S) in E. nigrum being approximately the same or higher near the smelter than further by chemical analyses (Uhlig et al.. 2000). Although the accumulation of metals was not clearly associated with the ele vated amounts of macronutrients. the localization of nutrients in different cell compartments 30 was in agreement with the function of these elements in the cells (e.g. Marschner. 1995: Keltjens and van Beusichem. 1998). No connection was found between the location of heavy metals and nutrients. The highest element (Ca. P. S. Al, As) peaks were found in the electron-dense material but these were not localized in the same tissue or distance. According to light microscopical examinations this electron-dense material consisted partly of phenolic material. Due to the higher fre quency of this phenolic, electron-dense material and the high heavy metal amounts meas ured by EDXS in the stems from the highly polluted site as well as the importance of phe nolics in the ecology of E. nigrum (e.g. Nilsson, 1992; Wallstedt. 1998). it can be assumed that this phenolic electron-dense material might have a function in the heavy metal toler ance of E. nigrum. Kukkola (1999) reported increased tannin production in pine needles, which was seen as dark staining of the central vacuole in the Cu and Ni treatments, and suggested that the dark staining may be caused by metal accumulation in the vacuoles (Kukkola. 1999). The increase in tannin amount in mesophyll cell vacuoles of pine needles, however, could be also due to the nutrient deficiencies (Kukkola et al., 1997) as shown by Holopainen and Nygren (1989). In Minuartia vema leaves Cu and Fe were found in the cell walls of parenchyma and Cu, Fe and Al in leaf surface, whereas the cell organelles did not contain any metals (Neumann et al., 1997). In this study, especially As was localized more frequently in the primary cell walls of living cells (ray cells, phloem) than dead cells, although it also accumulated in the cytoplasm and vacuoles. Fe was more frequent in the living (ray cells, phloem) than dead tissue (xylem) and Al did not show very clear pattern between the distances. Cu was not clearly distributed in certain cell organelles of stems. Also Neumann et al. (1995) reported that Cu was located in several cell organelles in leaves and roots of tolerant Armaria mari time! by electron microscope. Kukkola (1999) found that needles of CuNi irrigated pines in the arctic (North Finland) contained dark accumulation in cytoplasm and concluded that it might be partly due to metal accumulation as Cu is present in different deposits of the cells (Kukkola. 1999). Although the spectras showed differences in metal amounts in different cell compartments, the elemental mappings made in this study showed that Cu. Fe, As and Al were relatively 31 evenly distributed among the tissue. Because the heavy metals were not only localized in dead tissue, the heavy metal complexing agents could be involved in heavy metal tolerance of /'.', nigrum. However, heavy metal tolerance is metal-specific (Baker. 1981. 1987). and the specification of complexing agents of certain metals could not be done. Because of the ho mogeneous distribution of metals in the tissue, it seems that not only one mechanism is in volved in heavy metal tolerance of E. nigrum. Most of the heavy metals are present in the xvlem as free cations or complexed with organic acids, while Cu is mainly transported in the xylem in complexed form (Graham, 1979; White et al.. 1981; Kochian, 1991). Cu tolerance has been found to be achieved by chelating the Cu with proteins and vacuolar phenolic com pounds in leaves and roots (e.g. Rauser. 1984; Neumann et al., 1995). Cu and Fe accumula tion in C. vulgaris ericoid mycorrhizal roots has been suggested to be responsible for the re duction of Cu and Fe transport to the shoots when exposed to high Cu and Fe concentrations (e.g. Bradley et al.. 1981. 1982; Shaw et al.. 1990). In E. nigrum the considerable amounts of metals are transported also to the shoot. Several studies have shown the connection be tween Al tolerance and organic acids (e.g. Foy et al.. 1990; Larsen et al., 1998). Especially in crops the Al-induced organic acid release from roots is an important Al tolerance mechanism (Kochian and Garvin. 1999). Ahonen-Jonnarth (2000) also found that the oxalic acid production of I'm us svlvestris mycorrhizal and non-mycorrhizal roots increased when exposed to Al or Cu. 4.4. The morphological responses of C. vulgaris and E. nigrum to Cu and/or Ni in controlled conditions In this study, as well as in previous studies (e.g. Roth et al.. 1971; Taylor and Crowder. 1983; Kramer et al., 1996; L'Huillier et al., 1996). the growth parameters of plants were found to clearly respond to Cu and Ni. A clear decrease in the elongation and biomass of shoots and roots indicate the toxicity of elevated concentrations of Cu and/or Ni to C. vul garis and E. nigrum (I. II) shown also by ecophysiological parameters of E. nigrum near the Cu-Ni smelter at Harjavalta (IV). The growth decrease in greenhouse was higher than that measured in the field at Harjavalta (Salemaa et al.. 1995) and in the Kola Peninsula (Shcvtsova. 1998). Shevtsova (1998) reported 30% decrease in length growth and 65% de crease in biomass of E. nigrum in the Kola Peninsula compared to the unpolluted areas, where both increased heavy metal concentrations and SO2 affects on the growth of the 32 plants. In greenhouse, the maximum applied concentrations were high, because other factors (e.g. organic matter) that could affect the availability of metals in the soil-plant system were not present. In addition, at Harjavalta, the sulphuric acid and heavy metal load have strongly decreased since 1980' s (Rantalahti. 1995) and the pollutants enter the plant as dry or wet deposition. The most severe stress for plants, however, are the toxic metal concentrations in the soil. At Harjavalta, the total metal pool is not available for plants, because the Cu and Ni are in the soil partly in the complexed form (Derome and Lindroos, 1998b). At 0,5 km dis tance Cu is accumulated in the organic layer whereas Ni is less readily bound and more uniformly ditributed down the soil profile. This indicates that Cu is strongly retained in the organic layer and forms rather stable complexes with organic matter while Ni is more mo bile (Gorvainova and Nikonov. 1993) and therefore more plant available than Cu (Derome and Lindroos. 1998b). The relatively high decrease in growth already at the concentration of 10 mg l" 1 was surpris ing. because according to the pilot study, the growth of E. nigrum increased two times in the presence of 5 mg l" 1 Cu compared to that of 0 mg l" 1 Cu, indicating that E. nigrum would need some amounts of Cu for maximal growth (Monni and Salemaa. 1998). However, the pilot study was done earlier in the growing season, which might affect on the measured length growth of E. nigrum. 4.5. The heavy metal resistance of C. vulgaris and E. nigrum 4.5.1. Mechanisms As mentioned above. E. nigrum tolerated elevated concentrations of Cu and Ni in the stems, while transport to the green leaves was restricted (I). In the cellular level elevated As, Cu and Fe amounts were localized in cell walls, cytoplasm and vacuoles and phenolic electron dense material might also contribute to the heavy metal tolerance of E. nigrum (V). C. vul garis had higher Cu concentrations in discoloured leaves, stems and roots (II). According to the growth responses and survival. E. nigrum was more tolerant to Cu than C. vulgaris (I, II) as reported also by Laaksovirta and Silvola (1975). Also Gilbert (1975) reported that dy ing C. vulgaris was replaced by E. nigrum near aluminium smelters. 33 The greenhouse experiment partly helps to explain occurrence of E. nigrum in highly pol luted areas near the smelters (I). As mentioned above, according to Baker and Walker (1989), tolerant individuals can be separated from non-tolerant ones on the basis of their ability to establish, survive and reproduce in metal-contaminated substrates. E. nigrum is one of the surviving species near the smelters (e.g. Salemaa et al.. 2000b). It is capable of producing berries near the Cu-Ni smelter at Harjavalta and viable seeds occur at Harjavalta (personal observation) and in seedbank soil at Kola Peninsula (Komulainen et al., 1994) but, because of the toxic metal concentrations in the soil (Derome and Lindroos. 1998 a; Uhlig et al.. 2000). no new seedlings or vegetative reproduction have been observed. In natural habitats where the heavy metal concentrations are low. E. nigrum usually reproduces vege tativelv (Ojala. 1991). Not only the innate tolerance of E. nigrum, but also the heterogeneity of the contaminated soil may partly explain the successful survival of E. nigrum in Harjavalta and Kola Penin sula (Uhlig et al.. 2000). Uhlig et al.. (2000) reported that beneath E. nigrum patches (microsites) the concentrations of Cu. Ni. Fe. Pd. Zn. Al in the organic soil were considera bly higher, but those in the mineral soil considerably lower than the concentrations of the surrounding soil, where E. nigrum was not growing. They suggested that phenolics might have some role in immobilizing metals beneath E. nigrum. In polluted areas the taproot of E. nigrum penetrates in the mineral soil, and the living fine roots occur in the less polluted mineral soil. In the reference areas, in contrast, the fine roots lie in the organic soil. Wat mough and Dickinson (1995) discussed that the spatial heterogeneity of metal concentra tions and availability in surface soils might provide partial explanation for the survival of long-lived plants. Turner and Dickinson (1993) showed that the Acer pseudoplatanus L. (sycamore) did not tolerate metals although growing on the metal contaminated soils. They found that roots could grow in uncontaminated zones of the soils, and therefore seedlings survived in the contaminated soil. They concluded that the phenotypic plasticity might have an importance in long-lived plants (Turner and Dickinson, 1993). The environmental conditions may regulate the occurrence of C. vulgaris and explain why it grows in heavy metal polluted areas on oceanic heathlands but not in polluted boreal forests. In addition to high metal concentrations (Derome and Lindroos. 1998 a). drainage and a de ficiency of base cations near the Cu-Ni smelter at Harjavalta (Derome and Nieminen. 1998) 34 may explain the absence of C. vulgaris. The water supply is known to be critical for seedling establishment of C. vulgaris (e.g. Gimingham, 1972) (II). It has been suggested in earlier studies that the Cu resistance of C. vulgaris is based on an cricoid mycorrhiza. which accumulates the metals and thus prevents Cu transportation to the shoots of C. vulgaris (Bradley et al., 1981. 1982; Burt, 1984). Indications of true Cu tol erance in C. vulgaris have also been reported (Burt, 1984). The higher Cu concentrations in the roots than in the shoots found in this study support the hypothesis of Bradley et al. (1981. 1982) about the avoidance mechanism. Hashem (1987) compared the copper toler ance between the ectomycorrhizas and ericoid mycorrhiza Hymenoscyphus ericae coloniz ing the roots of C. vulgaris and found that Hymenoscyphus ericae was more tolerant than other species (Hashem. 1987). Shaw (1987). in contrast, researched the Al and Fe concen trations in the shoots and roots in mycorrhizal and non-mycorrhizal C. vulgaris seedlings, and found that the accumulation was metal-specific. Shoot Fe concentrations were signifi cantly higher in mycorrhizal than non-mycorrhizal plants in low and high Fe concentra tions. In contrast, Al concentrations in shoots and roots of mycorrhizal C. vulgaris seedlings were significantly lower than those in non-mycorrhizal seedlings (Shaw. 1987). Based on the high concentrations in the stems. Cu was also transported to the stems of C. vulgaris. The discoloured leaves had higher Cu concentrations than the green leaves, indi cating that in higher concentrations C. vulgaris could not restrict the Cu transport to the leaves. Marrs and Bannister (1978) and Burt (1984) concluded that C. vulgaris could de toxify and remove metals by accumulating them in older shedding leaves and stems (II). The results of this thesis, combined with the findings of earlier studies, provide a lot of new information about the heavy metal resistance of dwarf shrubs in the polluted area. However, although this study has answered many questions, many questions still remain open. Ac cording to the earlier field studies and this thesis, the specific tolerance and avoidance mechanisms of E. nigrum and C. vulgaris can be assumed to follow the pattern shown in Figure 3. especially concerning the heavy metal resistance of E. nigrum. 35 In addition to tolerance and avoidance mechanims the natural capability to adapt in chang ing environment and conditions where metal availability is high, is probably partly the rea son for the successful survival of E. nigrum in the polluted areas. Figure 3. The possible tolerance and avoidance mechanisms of E. nigrum and C. vulgaris based on this thesis and earlier studies. 4.5.2. The estimation of heavy metal resistance in the greenhouse and field The positive feature about greenhouse studies is that factors other than heavy metals can be eliminated, whereas in the field the observed responses may be also due to other environ mental factors. First of all. the greenhouse conditions are different from those in the field. 36 where the presence of organic matter, interactions with other species etc. affect metal avail ability for plants. Secondly, greenhouse experiments are of relatively short duration and therefore only give an approximation of what happens in the field. This is especially the case for long-lived plants although very short experimental periods to estimate the tolerance index by root growth have been used. Punshon and Dickinson (1997) showed that the accli mation of willows to toxic metals was achieved by the gradual increase of the toxic metals in the substrate rather than by a short-term greenhouse experiment. Therefore, the results in the greenhouse should always be compared with the situation in the field where the actual processes take place. 37 5. SUMMARY AND CONCLUSIONS Dwarf shrub species, besides growing naturally in nutrient-poor and dry conditions, seem to be resistant to heavy metals. The greenhouse studies confirmed the observations made in the field that C. vulgaris is more sensitive to Cu than E. nigrum. Toxicity symptoms also oc curred earlier in C. vulgaris than in E. nigrum. The exposure of C. vulgaris seedlings to Cu caused clear toxicity symptoms in the roots of the plants, and strongly affected survival. Toxicity symptoms also appeared on E. nigrum and growth was suppressed, but the metals did not affect survival during the six-week experiment (I. II). Both species were concluded to be resistant to elevated concentrations of Cu and/or Ni. be cause the concentrations in the plant parts were well above the level generally assumed to be toxic for plants (e.g. Allawav, 1968; Marschner, 1995). However, the difference in the con centrations between different plant parts was not as clear for C. vulgaris (II) as for E. nig rum (I). C. vulgaris accumulated high concentrations of Cu in its living parts, especially in the roots, stems and discoloured, shedding leaves during the short exposure period, and the accumulation pattern was similar throughout the concentration range (II). In E. nigrum. the Cu and Ni concentrations were the highest in the stems, and transport to the green leaves appeared to have been restricted, as reflected by the relatively constant Cu and Ni concen trations in the green leaves. The tolerance mechanism of E. nigrum was concluded to be partly achieved through the accumulation of metals in the older stems (I), and partly through the metal accumulation in cell walls, vacuoles and cytoplasm of stems in the cellu lar level. Heavy metal tolerance might also be achieved through the detoxification of metals in the electron-dense material of the phloem and ray cells. At the cellular level, it seems that not only one specific mechanism contribute to the heavy metal tolerance of E. nigrum (V). Alhough surviving near Cu-Ni smelters in the northern hemisphere (Helmisaari et ai.. 1995; Uhlig et ai.. 2000) and accumulating high concentrations of Cu and Ni (I), the vitality of E. nigrum near the smelters has clearly been decreased by the heavy metal load. This was ap parent from the lower organic acid and chlorophyll contents and higher ABA contents near the smelter (IV). However, the uptake of metals by the leaves seemed to be effectively re stricted. because the heavy metals applied to the aerial parts of E. nigrum in the greenhouse 38 were accumulated strongly on the bark and leaves of E. nigrum. Surface contamination had negative effects on photosynthesis, although it did not have any effect on the water potential and chlorophyll pigments. It decreased dark respiration and maximum photosynthesis and increased leaf ABA content (III). An ability to grow in heavy-metal polluted areas makes E. nigrum a suitable species for the remediation of contaminated soil. It is one of the plant species, which has been used in a revegetation experiment in the vicinity of the Harjavalta smelter (Kiikkilä et ai., 1996). C. vulgaris, on the other hand, was more sensitive to metals and is therefore not the most suit able species for the remediation of heavy metal polluted soil in the northern hemisphere. 39 6. REFERENCES Ahonen-Jonnarth. U., 2000. Growth, nutrient uptake and ectomycorrhizal function in Pi nus svlvestris plants exposed to aluminium and heavy metals. PhD thesis. Department of Forest Mycology and Pathology, Swedish University of Agricultural Sciences. Acta Universitatis Agriculturae Sueciae, Silvestria 130. Sweden. Allaway, W.H., 1968. Agronomic controls over the environmental cycling of trace elements Advances in Agronomy, 20, 235-274. Amiro. 8.D.. Courtin. G.M.. 1981. Patterns of vegetation in the vicinity of an industrially disturbed ecosystem. Sudbury. Ontario. Canadian Journal of Botany 59. 1623-1639. Angelov. M.. Tsonev, T.. Uzunova, A.. Gaidardjieva, K., 1993. Cu 2, effect upon photosyn thesis, chloroplast structure. RNA and protein synthesis of pea plants. Photosvn thetica 28. 341-350. Antonovics, J., Bradshaw, A.D.. Turner, R.G.. 1971. Heavy metal tolerance in plants. Advances in Ecological Research 7, 1-85. Bagatto. G., Shorthouse. J.D.. 1991. Accumulation of copper and nickel in plant tissues and an insect gall of lowbush blueberry, Vaccinium angustifolium, near an ore smelter at Sudbury. Ontario. Canada. Canadian Journal of Botany 69. 1483-1490. Bagatto. G.. Crowder. A.A., Shorthouse. J.D.. 1993. Concentrations of metals in tissues of Lowbush Blueberry (Vaccinium angustifolium) near a copper-nickel smelter at Sudbury. Ontario. Canada: A factor analytic approach. Bulletin of Environmental Contamination and Toxicology 51. 600-604. Bagatto. G., Shorthouse, J.D.. 1999. Biotic and abiotic characteristics of ecosystems on acid metalliferous mine tailings near Sudbury. Ontario. Canadian Journal of Botany 77, 410-425. Bajwa, R.. Abuarghub, S., Read. D.J.. 1985. The biology of mycorrhiza in the Ericaceae. X The utilization of proteins and the production of proteolytic enzymes by the mycorrhizal endophyte and by mycorrhizal plants. New Phytologist 101. 469-486. Baker. A.J.M.. 1981. Accumulators and excluders - strategies in the response of plants to heavy metals. Journal of Plant Nutrition 3. 643-654. Baker, A.J.M., 1987. Metal tolerance. New Phytologist 106, 93-111. Baker, A.J.M., Walker. P.L.. 1989. Physiological responses of plants to heavy metals and the quantification of tolerance and toxicity. Review. Chemical Speciation and Bioavailabilitv 1. 7-17. Baker. A.J.M.. Walker. P.L.. 1990. Ecophysiologv of metal uptake by tolerant plants. In: Shaw. A.J., (Ed.). Heavy metal tolerance in plants: Evolutionary aspects, pp. 155- 177. CRC Press. FL, USA. Balsberg Pählsson. A.-M., 1989. Toxicity of heavy metals (Zn, Cu, Cd. Pb) to vascular plants. A literature review. Water. Air and Soil Pollution 47. 287-319. Beardsell, M.F., Cohen. D., 1975. Relationships between leaf water status, abscisic acid levels and stomatal resistance in maize and sorghum. Plant Physiology 56, 207- 212. Becernl. J.M.. Conzälez-Munia. C.. Munoz-Rueda. A.. De Felipe. M.R.. 1989. Changes induced by cadmium and lead in gas exchange and water relations of clover and lucerne. Plant Physiology and Biochemistry 27, 913-918. Beckett. P.H.T.. Davis. R.D., 1977. Upper critical levels of toxic elements in plants. New Phytologist 79. 95-106. 40 Bell, J.N.8., 1969. The autecology of Empetrum nigrum L., with particular reference to its recent behaviour at Wybunbury moss, Cheshire. PhD thesis. The Victoria Univer sity of Manchester, UK. Bell. J.N.8.. Tallis. J.H., 1973. Biological flora of the British Isles. Empetrum nigrum L Journal of Ecology 61. 289-305. Bishnoi, N.R., Sheoran. 1.5.. Singh. R., 1993. Influence of cadmium and nickel on photosynthesis and water relations in wheat leaves of different insertion level Photosynthetica 28, 473-479. Boehringer, 1989. Methods of biochemical analysis and food analysis using single reagents. Boehringer Mannheim GMBH, Biochemica, Mannheim, Germany Boyd. R.S., Jaffre. T., Odom. J.W.. 1999. Variation in nickel content in the nickel hvperaccumulating shrub Psychotria douarrei (Rubiaceae) from New Caledonia Biotropica 31. 403-410. Bradley. R.. Burt. A.J., Read. D.J., 1981. Mycorrhizal infection and resistance to heavy metal toxicity in Calluna vulgaris. Nature 292. 335-337. Bradley, R., Burt. A.J., Read. D.J.. 1982. The biology of mycorrhiza in the Ericaceae. VIII The role of mycorrhizal infection in heavy metal resistance. New Phvtologist 91, 197-209. Bradshaw, A.D.. Hardwick, K.. 1989. Evolution and stress - genotypic and phenotypic components. Biological Journal of the Linnean Society 37, 137-155. Burt. A. J., 1984. Ecophysiological studies of heavy metal resistance in Calluna vulgaris (L.) Hull, with special reference to mycorrhizal infection. PhD thesis, Department of Botany, University of Sheffield. UK. Burton, K.W., Morgan, E., 1983. The influence of heavy metals upon the growth of sitka spruce in South Wales forests. I. Upper critical and foliar concentrations. Plant and Soil 73. 327-336. Bucking. H.. Beckmann, S.. Heyser. W. and Kottke. 1., 1998. Elemental contents in vacuolar granules of ectomycorrhizal fungi measured by EELS and EDXS. A comparison of different methods and preparation techniques. Micron 29, 53-61 Carlquist, S., 1989. Wood and bark anatomy of Empetraceae; comments on paedomorphosis in woods of certain small shrubs. Aliso 12, 497-515. Carroll. J. A., Caporn, S.J.M., Cawley, L., Read. D.J., Lee, J. A., 1999. The effect of in creased deposition of atmospheric nitrogen on Calluna vulgaris in upland Britain. New Phvtologist 141, 423-431. Chertov, 0.G., Nadporozshskaya. M.A., Lapshina, 1.N., Grigorieva, 0.A., 1993. Soil degradation in surroundings of "Pechenganikel" smelter complex. In: Kozlov. M. V., Haukioja, E., Yarmishko. V.T.. (Eds), Aerial pollution in Kola Peninsula. Proceedings of the international workshop, 14-16 April, St Petersburg, pp. 159- 162. Claussen, W., Lenz, F., 1999. Effect of ammonium or nitrate nutrition on net photosyn thesis. growth and activity of the enzymes nitrate reductase and glutamine syn thetase in blueberry, raspberry and strawberry. Plant and Soil 208. 95-102. Davies, W.J., Zhang, J., 1991. Root signals and the regulation of growth and development of plants in drying soil. Annual Review of Plant Physiology and Plant Molecular Biology 42. 55-76. Davis. R.D., Beckett, P.H.T., 1978. Upper critical levels of toxic elements in plants. 11. Critical levels of copper in young barley, wheat, rape, lettuce and ryegrass, and of nickel and zinc in young barley and ryegrass. New Phvtologist 80. 23-32. 41 Delfine. S.. Alvino. A., Villani. M.C.. Loreto. F.. 1999. Restrictions to carbon dioxide conductance and photosynthesis in spinach leaves recovering from salt stress. Plant Physiology 119. 1101-1106. Derome. J.. 2000. Effects of heavy-metal and sulphur deposition on the chemical proper ties of forest soil in the vicinity of a Cu-Ni smelter, and means of reducing the detrimental effects of heavy metals. PhD thesis. Finnish Forest Research Institute. Research Papers 769. Finland. Derome. J., Lindroos. A.-J., 1998 a. Effects of heavy metal contamination on macronutri ent availability and acidification parameters in forest soil in the vicinity of the Harjavalta Cu-Ni smelter. SW Finland. Environmental Pollution 99. 225-232. Derome. J.. Lindroos, A.-J., 1998b. Copper and nickel mobility in podzolic forest soils subjected to heavy metal and sulphur deposition in Western Finland. Chemosphere 36,1131-1136. Derome. J.. Nieminen, T.. 1998. Metal and macronutrient fluxes in heavy-metal polluted Scots pine ecosystems in SW Finland. Environmental Pollution 103. 219-228. Dickinson. N.M.. Turner, A.P.. Lepp, N.W.. 1991. How do trees and other long-lived plants survive in polluted environments. Functional Ecology 5, 5-11. Eltrop, L.. Brown. G.. Joachim. 0., Brinkmann, K.. 1991. Lead tolerance of Be tula and Salix in the mining area of Mechernich/Germany. Plant and Soil 131. 275-285 Elvebakk. A., Spjelkavik. S.. 1995. The ecology and distribution of Empetrum nigrum ssp hermaphroditum on Svalbard and Jan Mayen. Nordic Journal of Botany 15. 541- 552. Ernst, W.H.0., 1975. Physiology of heavy metal resistance in plants. Proceedings of the international conference of heavy metals in the environment. Toronto, 2. pp. 121- 136. Ernst. W.H.0., Verkleij. J.A.C.. Schat, H.. 1992. Metal tolerance in plants. Acta Botanica Neerlandica 41. 229-248. Eurola. S.. Bendiksen. K., Rönkä. A.. 1990. Suokasviopas. Oulanka Reports 9. University of Oulu. 205 p. Fov. C.D.. Lee, E.H., Wilding, 5.8., 1987. Differential aluminium tolerances of two barley cultivars related to organic acids in their roots. Journal of Plant Nutrition 10. 1089- 1101. Fov. C.D., Lee, E.H.. Coradetti. C.A., Taylor. G.J.. 1990. Organic acids related to differential aluminium tolerance in wheat (Triticum aestivum) cultivars. In: van Beusichem. M.L. (Ed.), Plant nutrition - physiology and applications. 381-389. Freedman. 8.. Hutchinson, T.C.. 1980. Pollutant inputs from the atmosphere and accu mulations in soils and vegetation near a nickel-copper smelter at Sudbury, Ontario. Canada. Canadian Journal of Botany 58. 108-132. Fritz, E.. 1980. Microradioautographic localization of assimilates in phloem: problems and new method. Berichte der Deutschen Botanischen Gesellschafit 93, 109-121. Fritze, H.. Vanhala. P., Pietikäinen. J., Mälkönen. E., 1996. Vitality fertilization of Scots pine stands growing along a gradient of heavy metal pollution: short-term effects on microbial biomass and respiration rate of the humus layer. Fresenius Journal of Analytical Chemistry 354, 750-755. Gerdol, R.. Brancaleoni, L., Menghini. M.. Marchesini. R.. 2000 a. Response of dwarf shrubs to neighbour removal and nutrient addition and their influence on com munity structure in a subalpine heath. Journal of Ecology 88. 256-266. 42 Gerdol, R., lacumin, P., Marchesini, R.. Bragazza, L., 2000b. Water- and nutrient-use efficiency of a decidious species, Vaccinium myrtillus, and an evergreen species, Ivitis-idaea, in a subalpine dwarf shrub heath in the southern Alps. Italy. Oikos 88, 19-32. Gilbert. 0.L., 1975. Effects of air pollution on landscape and land-use around Norwegian aluminium smelters. Environmental Pollution 8, 113-121. Gimingham, C.H.. 1960. Biological flora of the British isles. Calluna vulgaris (L.) Hull Journal of Ecology 48, 455-483. Gimingham. C.H., 1972. Ecology of heathlands. pp. 87-109. John Wiley & Sons, Inc. New York. USA. Godbold, D.L.. Horst, W.J.. Collins, J.C.. Thurman. D.A., Marschner. H.. 1984. Accumula tion of zinc and organic acids in roots of zinc tolerant and non-tolerant ecotypes of Deschampsia caespitosa. Journal of Plant Physiology 116, 59-69. Good. R. D'O.. 1927. The genus Empetrum L. Journal of the Linnean Society of London 47. 489-523. Gorvainova. V.. Nikonov. V.. 1993. Adsorption capacity of podzolic Al-Fe-humus soils under pine with respect to copper and nickel. In: Derome. J. (Ed.). Russian- Finnish cooperation report. Mimeograph, Rovaniemi Research Station. Finnish Forest Research Institute, Finland, 94-97. Graan. T.. Ort. D R.. 1984. Quantitation of the rapid electron donors to P7OO. the functional plastoquinone pool, and the ratio of the photosystems in spinach chloroplasts. The Journal of Biological Chemistry 259, 14003-14010. Graham, R.D.. 1979. Transport of copper and manganese to the xylem exudate of sunflo wer. Plant, Cell and Environment 2, 139-143. Gregory. R.P.G.. Bradshaw. A.D., 1965. Heavy metal tolerance in populations of Agrostis tenuis Sibth. and other grasses. New Philologist 64. 131-143. Grubb, P.J.. Suter, M.8.. 1971. The mechanism of acidification of soil by Calluna and IJ lex and the significance for conservation. In: Duffev. E.. Watt. A.S.. (Eds). The scientific management of animal and plant communities for conservation, pp. 115- 133. Blackwell Sei. Publ. Oxford. London, Edinburgh, UK. Haapasaari, M.. 1988. The oligotrophic heath vegetation of northern Fennoscandia and its zonation. Actaßotanica Fennica 135, 1-219. Hampp, R., Goller, M., Ftillgraf, H.. 1984. Determination of compartmented metabolite pools by a combination of rapid fractionation of oat mesophyll protoplasts and en zvmic cycling. Plant Physiology 75, 1017-1021. Hartley, S.E.. Amos. L.. 1999. Competitive interactions between Nardus stricta L. and Cal luna vulgaris (L.) Hull: the effect of fertilizer and defoliation on above- and below ground performance. Journal of Ecology 87, 330-340. Hashem, A.R., 1987. The role of mycorrhizas in the resistance of plants to metals. PhD the sis. Department of Botany, University of Sheffield. UK. Haslam. E.. 1977. Symmetry and promiscuity in procyanidin biochemistrv. Photochemistry 16. 1625-1640. Helander. M.L. 1995. Responses of pine needle endophvtes to air pollution. New Phytologist 131. 223-229. Helmisaari. H.-S.. Derome. J.. Levula, T., Nieminen, T., Palmgren. K., Salemaa. M.. Vanha-Majamaa. 1.. Reponen. P.. Rutter. 0.. 1994. Metsikön ravinnekierto ilman epäpuhtauksien aiheuttaman kuormituksen osoittajana. In: Mälkönen. E.. Sivula. H.,(Eds.). Suomen metsien kunto. Metsäntutkimuslaitoksen tiedonantoja 527. pp. 156-174. 43 Helmisaari. H.-S.. Derome. J.. Fritze, H.. Nieminen, T., Palmgren, K.. Salemaa. M., Vanha- Majamaa, 1., 1995. Copper in Scots pine forests around a heavy-metal smelter in South-Western Finland. Water, Air and Soil Pollution 85, 1727-1732. Hicklenton. P.R., Reekie, J.Y., Gordon, R.J., Percival. D.C., 2000. Seasonal patterns of photosynthesis and stomatal conductance in lowbush blueberry plants managed in a two-year production cycle. HortScience 35, 55-59. Holopainen, T.. Nygren, P.. 1989. Effects of potassium deficiency and simulated acid rain alone and in combination on the ultrastructure of Scots pine needles. Canadian Journal of Forest Research 19. 1402-1411. Hoogesteger, J.. Karlsson. P.S., 1992. Effects of defoliation on radial stem growth and photosynthesis in the mountain birch (Betula pubescens ssp. tortuosa). Functional Ecology 6. 317-323. lason, G.R.. Hartley, S.E.. Duncan. A.J.. 1993. Chemical composition of Calluna vulgaris (Ericaceae): does response to fertiliser vary with phenological stage? Biochemical Systematics and Ecology 21. 315-321. Jalal. M.A.F., Read, D.J., 1983. The organic acid composition of Calluna heathland soil with special reference to phyto- and fungitoxicity. 11. Monthly quantitative determination of the organic acid content of Calluna and spruce dominated soils. Plant and Soil 70, 273-286. Johanson, U.. Gehrke, C.. Björn, L.0., Callaghan, T.V., Sonesson, M., 1995. The effects of enhanced UV-B radiation on a subarctic heath ecosystem. Ambio 24, 106-111. Johnsen, 1.. Ro-Poulsen. H., Sochting, U., Mortensen. L.. 1991. Gasformige luftforurening ers effekter pä danske plantesamfund. Rapport, Energiministeriet, 1-31. Keltjens. W.G. and van Beusichem, M.L.. 1998. Phytochelatins as biomarkers for heavy metal stress in maize (Zea mays L.) and wheat (Triticum aestivum L.): combined effects of copper and cadmium. Plant and Soil 203. 119-126. Kiikkilä. 0.. Uhlig, C.. Helmisaari, H.-S., Salemaa, M., Fritze. H., Nieminen. T., Vanha- Majamaa. 1.. Derome. J. 1996. Revegetation experiment in heavy metal contami nated Scots pine forest at Harjavalta, western Finland. In: Kopponen P., Kären lampi. S.. Rekilä. R.. Kärenlampi, L., (Eds), Bio- and Ecotechnological Methods in Restoration. Abstracts of an International Advanced Course and Minisymposium. 16-18 December. Kuopio, Finland, pp. 9. Kochian, L.V.. 1991. Mechanisms of micronutrient uptake and translocation in plants. In Mortvedt, J.J. (ed.). Micronutrients in agriculture, pp. 229-296. Soil Science Society of American Book Series. No 4. Madison. WI. USA. Kochian. L.. Garvin, D.. 1999. A physiological and genetic investigation of the role of root exudates in aluminum tolerance. In: Wenzel, W.W.. Adriano. D.C.. Allo wav. 8.. Doner. H E., Keller. C., Lepp, N.W.. Mench, M., Naidu, R.. Pierzynski. G.M. (Eds). Proceedings of extended abstracts. sth international conference on the biogeochemistry of trace elements. July 11-15. Vienna, Austria, pp. 158-159. Komulainen. M., Vieno, M.. Yarmishko. V.T., Daletskaja. T.D., Maznaja, E.A., 1994. Seedling establishment from seeds and seed banks in forests under long-term pollution stress: a potential for vegetation recovery. Canadian Journal of Botany 72. 143-149. Kottke. 1.. 1994. Localization and identification of elements in mycorrhizas. Advantages and limits of electron energy loss spectroscopy. Acta Botanica Gallica 141. 507- 510. Kramer. U.. Cotter-Howells. J.D.. Charnock, J.M.. Baker, A.J.M., Smith. J.A.C. 1996. Free histidine as a metal chelator in plants that accumulate nickel. Nature 379, 635-638. 44 Kukkola, E., 1999. Effects of copper and nickel on subarctic Scots pine needles. Acta Universitatis Ouluensis. A 336. PhD thesis. University of Oulu, Finland. Kukkola, E., Huttunen, S., Bäck, J., Rautio, P. 1997. Scots pine needle injuries at subarctic industrial sites. Trees 11, 378-387. Laaksovirta, K., Silvola, J., 1975. Effect of air pollution by copper, sulphuric acid and fertilizer factories on plants at Haijavalta. W. Finland. Annales Botanici Fennici 12, 81-88. Laine. J.. Vasander, H.. 1990. Suotyypit. Kirjayhtymä, Helsinki. 80 p Lamoreux. R.J., Chaney, W.R., 1978. The effect of cadmium on net photosythesis. transpi ration, and dark respiration of excised silver maple leaves. Physiologia Plantarum 43. 231-236. Larsen. P. 8.. Degenhardt. J., Tai. C.-Y.. Stenzler. L.M.. Howell. S.H.. Kochian. L.V., 1998. Aluminium-resistant Arabidopsis mutants that exhibit altered patterns of aluminium accumulation and organic acid release from roots. Plant Physiology 117, 9-18. Leake, J.R., 1988. Causes and effects of soil acidification by Calluna vulgaris (L.) Hull, with special reference to the role of mycorrhizas. PhD thesis. University of Sheffield. UK. Leake. J.R., Read, D.J., 1989. The biology of mycorrhiza in the Ericaceae. XIII. Some characteristics of the extracellular proteinase activity of the ericoid endophyte Hymenoscyphus ericae. New Phytologist 112, 69-76. Lee, J.. Reeves, R.D.. Brooks. R.R., Jaffre T., 1978. The relation between nickel and citric acid in some nickel-accumulating plants. Phytochemistry 17. 1033-1035 L'Huillier, L.. d'Auzac. J.. Durand. M.. Michaud-Ferriere. N., 1996. Nickel effects on two maize (Zea mays) cultivars: growth, structure, Ni concentration, and localiza tion.Canadian Journal of Botany 74, 1547-1554. Lichtenberger. O. and Neumann. D., 1997. Analytical electron microscopy as a powerful tool in plant cell biology: examples using electron energy loss spectroscopy and x-ray microanalysis. European Journal of Cell Biology 73, 378-386. Lukina, N.V.. Liseenko. L.A., Belova, E.A., 1993. Pollution-induced changes in the vegetation cover of spruce and pine ecosystems in the Kola North region. In: Kozlov, M. V., Haukioja. E.. Yarmishko. V.T.. (Eds)., Aerial pollution in Kola Peninsula. Proceedings of the international workshop. 14-16 April. St. Petersburg, pp. 312-321. Macnair. M.R., 1990. The genetics of metal tolerance in natural populations. Heavy metal tolerance in plants: Evolutionary aspects. Shaw. A.J. (Ed.), pp. 235-253. CRC Press. FL. USA. Mallik. A.U., Hobbs, R.J.. Legg, C.J., 1984. Seed dynamics in Calluna-Arctostaphylos heath in north-eastern Scotland. Journal of Ecology 72, 855-871. Mallik. A.U.. Hobbs. R.J.. Rahman. A. A.. 1988. Seed-bed substrates and revegetation of Calluna heathlands following burning. Journal of Environmental Management 27 379-397. Maimer, N., Nihlgärd, 8.. 1980. Supply and transport of mineral nutrients in a sub-arctic mire. In: Sonesson, M. (Ed.), Ecology of a Subarctic Mire. Ecological Bulletins 30, 63-95. Marrs, R.H., Bannister, P., 1978. The adaptation of Calluna vulgaris (L.) Hull to contrast ing soil types. New Phytologist 81, 753-761. Marschner. H., 1995. Mineral nutrition of higher plants. 2nd edn. London, UK: Academic Press Ltd. 45 Mathys. W., 1977. The role of malate, oxalate and mustard oil glucosides in the evolution of zinc-resistance in herbage plants. Phvsiologia Plantarum 40, 130-136. Meharg, A.A.. Cairney, J.W.G., 2000. Co-evolution of mycorrhizal symbionts and their hosts to metal-contaminated environments. Advances in Ecological Research 30, 70-112. Michelsen. A., Jonasson, S., Sleep, D., Havström, M., Callaghan, T.V., 1996 a. Shoot biomass, S I3 C. nitrogen and chlorophyll responses of two arctic dwarf shrubs to in situ shading, nutrient application and warming simulating climatic change. Oecologia 105. 1-12. Michelsen, A., Schmidt, 1.K.. Jonasson, S., Quarmby, C., Sleep, D. 1996b. Leaf 15 N abun dance of subarctic plants provides field evidence that ericoid. ectomvcorrhizal and non- and arbuscular mycorrhizal species access different sources of soil nitrogen. Oecologia 105. 53-63. Miller, H.J., 1975. Anatomical characteristics of some woody plants of the angmagssalik district of southeast Greenland. Meddelelser om Gronland 198. 1-30. Mohamed, 8.F.. Gimingham. C.H., 1970. The morphology of vegetative regeneration in Calluna vulgaris. New Phytologist 69, 743-750. Monni, S., and Salemaa, M., 1998. Kanervan ja variksenmarjan herkkyys kuparille. In: Mälkönen, E. (toim.). Ympäristömuutos ja metsien kunto. Metsien terveydentilan tutkimusohjelman loppuraportti. Metsäntutkimuslaitoksen tiedonantoja 691. 196. Mullins, M.. Hardwick. K. and Thurman, D. 1985. Heavy metal location by analytical electron microscopy in conventionally fixed and freeze-substituted roots of metal tolerant and non tolerant ecotypes. In: Proceedings of international conference on heavy metals in the environment. Athens, pp. 43-46. Mälkönen. E„ Derome. J., Fritze, H., Helmisaari, H.-S., Kukkola, M., Kytö, M., Saar salmi, A., Salemaa, M., 1999. Compensatory fertilization of Scots pine stands polluted by heavy metals. Nutrient Cycling in Agroecosystems 55, 239-268. Nelson. D.W.. Sommers. L.E.. 1982. Total carbon, organic carbon and organic matter. In: Page et al.. (Eds.). Methods of soil analysis, part 11. pp. 574-577. 2. Edn. Soil Sci ence Society of America, Madison, Wisconsis, USA. Neumann, D.. zur Nieden, U., Lichtenberger, O. and Leopold, 1., 1995. How does Armeria maritima tolerate high heavy metal concentrations. Journal of Plant Physiology 146, 704-717. Neumann, D.. zur Nieden, U., Schwieger, W., Leopold, I. and Lichtenberger, 0., 1997. Heavy metal tolerance of Minuartia verna. Journal of Plant Physiology 151, 101 - 108. Nieminen, T.M., Derome. J., Helmisaari, H.-S., 1999. Interactions between precipitation and Scots pine canopies along a heavy-metal pollution gradient. Environmental Pollution 106. 129-137. Nilsson. M.-C.. 1992. The mechanisms of biological interference by Empetrum hermaph roditum on tree seedling establishment in boreal forest ecosystems. PhD thesis. Swedish University of Agricultural Sciences. Umea. Sweden. Nilsson. M.-C., 1994. Separation of allelopathy and resource competition by the boreal dwarf shrub Empetrum hermaphroditum Hagerup. Oecologia 98. 1-7. Nilsson. M.-C.. Zackrisson. 0.. 1992. Inhibition of Scots pine seedling establishment by Empetrum hermaphroditum. Journal of Chemical Ecology 18, 1857-1870. Nilsson. M.-C., Gallet, C., Wallstedt, A., 1998. Temporal variability of phenolics and batatasin-111 Empetrum hermaphroditum leaves over an eight-year period: interpretations of ecological fiinction. Oikos 81. 6-16. 46 Ojala. A.. 1991. Pohjanvariksenmarjan lisääntymisstrategiasta. Luonnon Tutkija 95. 53-55 Pandolfini. T„ Gabbrielli. R.. Ciscato. M., 1996. Nickel toxicity in two Durum wheat culti vars differing in drought sensitivity. Journal of Plant Nutrition 19, 1611-1627. Pellissier. F., Souto, X.C.. 1999. Allelopathy in Northern temperate and boreal semi natural woodland. Critical Reviews in Plant Sciences 18. 637-652. Pennanen. T., Frostegärd. Ä.. Frtze, H., Baath. E., 1996. Phospholipid fatty acid composi tion and heavy metal tolerance of soil microbial communities along two heavy metal-polluted gradients in coniferous forests. Applied and Environmental Micro biology 62. 420-428. Poschenrieder. C., Gunse. 8., Barcelö. J.. 1989. Influence of cadmium on water relations, stomatal resistance, and abscisic acid content in expanding bean leaves. Plant Physiology 90, 1365-1371. Proctor. J., Woodell. S.R.J., 1971. The plant ecology of serpentine. I. Serpentine vegetation of England and Scotland. Journal of Ecology 59, 375-395. Punshon, T.. Dickinson. N.M., 1997. Acclimation of Salix to metal stress. New Phvtologist 137. 303-314. Radin. J. W., 1984. Stomatal responses to water stress and to abscisic acid in phosphorus deficient cotton plants. Plant Physiology 76, 392-394. Rantalahti. R.. 1995. Outokummun Harjavallan tehtaiden ympäristönsuojelusta. Vasken viesti 3+4, 2-6. Rauser. W.E., 1984. Copper-binding protein and copper tolerance in Agrostis gigantea. Plant Science Letters 33, 239-247. Rauser. W.E.. Dumbroff, E. 8., 1981. Effects of excess cobalt, nickel and zinc on the water relations of Phaseolus vulgaris. Environmental and Experimental Botany 21, 249- 255. Read, D.J., 1983. The biology of mycorrhiza in the Ericales. Canadian Journal of Botany 61. 985-1004. Reimann, C.. Halleraker. J.H., Kashulina. G.. Bogatyrev. 1.. 1999. Comparison of plant and precipitation chemistry in catchments with different levels of pollution on the Kola Peninsula. Russia. The Science of the Total Environment 243/244, 169-191. Richter. H. 1997. Water relations of plants in the field: some comments on the measurement of selected parameters. Journal of Experimental Botany 48, 1-7. Ritchie. G.A., Hinckley, T.M., 1975. The pressure chamber as an instrument for ecological research. In: Macfadyen. A.. (Ed.), Advances in Ecological Research 9, 165-254. Robinson, R.K., 1971. Importance of soil toxicity in relation to the stability of plant com munities. In: Duffey E, Watt AS, eds. The scientific management of animal and plant communities for conservation, pp. 105-113. Blackwell Sei. Publ, Oxford. London. Edinburgh. UK. Roth, J.A., Wallihan, E.F., Sharpless, R.G., 1971. Uptake by oats and soybeans of copper and nickel added to a peat soil. Soil Science 112, 338-342. Salemaa, M., Ojala, A.. Vanha-Majamaa, 1., Vieno, M., 1995. Life-history of the crow berry Empetrum nigrum L. in heavy metal polluted environment. In: Oborny. 8.. De Kroon. H.. van Groenendael. J. (Eds.), The role of clonality in plant communi ties. Abstracts of the 4th international workshop on clonal plants. Visegard. Hungary. April 18-22, p. 46. Salemaa, M., Vanha-Majamaa. 1., 1998. Metsäkasvillisuus raskasmetallien saastuttamassa ympäristössä. In: Mälkönen, E. (Ed.), Ympäristönmuutos ja metsien kunto. Metsi en terveydentilan tutkimusohjelman loppuraportti. Metsäntutkimuslaitoksen tiedon antoja 691. 192-202. 47 Salemaa. M., Vanha-Majamaa, 1., Gardner, P.J., 1999. Compensatory growth of two clonal dwarf shrubs, Arctostaphylos uva-ursi and Vaccinium uliginosum in a heavy metal polluted environment. Plant Ecology 141, 79-91. Salemaa. M., Vanha-Majamaa, 1., Reinikainen, A., Nousiainen, H.. 2000 a. Responses of understorey vegetation to heavy metal loading. In: Forest condition in a changing environment. -The Finnish case. Mälkönen, E. (Ed), pp. 266-275. Salemaa. M., Vanha-Majamaa, 1.. Derome. J., 2000b. Understorey vegetation along a heavy metal pollution gradient in SW Finland. Environmental Pollution (in press). Sarvas. R., 1937. Havaintoja kasvillisuuden kehityksestä Pohjois-Suomen kuloaloilla. Silva Fennica 44. 1-64. SAS Institute Inc.. 1994. SAS/STAT user's guide. Version 6. 4th edn. Cary, North Caro lina, USA: SAS Institute Inc. Scholander. P.F.. Hammel. H.T., Bradstreet, E.D.. Hemmingsen, E.A., 1965. Sap pressure in vascular plants. Science 148, 339-346. Shaw, G., 1987. Iron and aluminium toxicity in the Ericaceae in relation to mycorrhizal infection. PhD thesis. Department of Botany, University of Sheffield, UK. Shaw. G., Leake, J.R., Baker, A.J.M., Read, D.J., 1990. The biology of mycorrhiza in the Ericaceae XVTI. The role of mycorrhizal infection in the regulation of iron uptake by ericaceous plants. New Phytologist 115, 251-258. Slicvtsova. A.. 1998. Responses of subarctic dwarf shrubs to climate change and air pollu tion. Annales Universitatis Turkuensis Ser. All. 113. PhD thesis. University of Turku. Finland. Shorthouse. J.D.. Bagatto. G.. 1995. Potential role of lowbush blueberry (Vaccinium an gustifoliuin) in colonizing metal-contaminated ecosystems. In: Gunn, J.M. (Ed.), Restoration and recovery of an industrial region. Progress in restoring the smelter damaged landscape near Sudbury, Canada, pp. 247-255. Springer-Verlag, New York. Berlin. Heidelberg. Sokal, R.R., Rohlf. F.J., 1995. Biometry. W. H. Freeman and Co., New York. 1-887 Spurr, A., 1969. A low-viscosity epoxy resin embedding medium for electron microscopy Journal of Ultrastructural Research 26, 31-43. Statistix. 1996. Statistix for windows. User's manual. Borland International Inc.. Tallahassee. FL, USA. Stelzer. R. and Lehmann. H., 1993. Recent developments in electron microscopical tech niques for studying ion localization in plant cells. Plant and Soil 155/156, 33-43 Strandberg, M., Johansson. M., 1999. Uptake of nutrients in Calluna vulgaris seed plants grown with and without mycorrhiza. Forest Ecology and Management 114, 129- 135. Striblev, D.P.. Read. D.J., 1976. The biology of mycorrhiza in the Ericaceae. VI. The ef fects of mycorrhizal infection and concentration of ammonium nitrogen on growth of cranberry (Vaccinium microcarpon Ait.) in sand culture. New Phytologist 77, 63-72. Taylor. G. J.. Crowder. A. A., 1983. Uptake and accumulation of copper, nickel, and iron by Typha latifolia grown in solution culture. Canadian Journal of Botany 61.1825- 1830. Timmer, V.R.. 1991. Interpretation of seedling analysis and visual symptoms. In: Van den Driessche. R. (Ed.). Mineral nutrition of conifer seedlings, pp. 113-134. CRC Press. Boca Raton. 274 p. Tomsett. A.8., Thurman, D.A., 1988. Molecular biology of metal tolerances of plants Plant Cell and Environment 11. 383-394. 48 Turnau, K.. Kottke, I. and Oberwinkler. F., 1993 a. Element localization in mycorrhizal roots of Pteridium aquilinum (L.) Kuhn collected from experimental plots treated with cadmium dust. New Phytologist 123. 313-324. Turnau. K., Kottke, I. and Oberwinkler. F., 1993b. Paxillus involutus-Pinus sylvestris mycorrhizae from heavily polluted forest. I. Element Localization Using Electron Energy Loss Spectroscopy and Imaging. Botanica Acta 106, 213-219. Turner. A.P., Dickinson, N.M., 1993. Survival of Acer pseudoplatanusL. (sycamore) seedlings on metalliferous soils. New Phytologist 123, 509-521. Tybirk, K, Nilsson, M.-C., Michelsen, A., Lakkenborg Kristensen, H., Shevtsova. A., Strandberg. M.T., Johansson. M.. Nielsen K.E., Riis-Nielsen, T., Strandberg, 8., Johnsen, 1., 2000. Nordic Empetrum dominated ecosystems: Function and susceptibility to environmental changes. Ambio 29, 90-97. Uhlig, C., Salemaa. M., Vanha-Majamaa, 1.. Derome, J., 2000. Element distribution in Empetrum nigrum microsites at heavy-metal contaminated sites in Harjavalta, Western Finland. Environmental Pollution (in press). Verkleij. J.A.C. and Schat. H.. 1990. Mechanisms of metal tolerance in higher plants. In: Shaw, A.J. (ed.), Heavy metal tolerance in plants: Evolutionary aspects, pp. 179- 193. CRC Press, FL. USA. Väisänen. S., 1986. Effects of air pollution by metal, chemical and fertilizer plants on forest vegetation at Kokkola, W Finland. Annales Botanici Fennici 23, 305-315. Wallstedt. A., 1998. Temporal variation and phototoxicity of Batatasin-111 produced by Empetrum hermaphroditum. PhD thesis. Department of Ecology. Chemical Ecology and Ecotoxicology, Lund University, Sweden. Watmough. S.A.. Dickinson. N.M.. 1995. Dispersal and mobility of heavy metals in re lation to tree survival in an aerially contaminated woodland soil. Environmental Pollution 90, 135-142. White, M.C.. Decker, A.M.. Chaney, R.L., 1981. Metal complexation in xylem fluid. I. Chemical composition of tomato and soybean stem exudate. Plant Physiology 67, 292-300. Wilkins. D.A., 1978. The measurement of tolerance to edaphic factors by means of root growth. New Phytologist 80, 623-633. Winterhalder, K., 1995. Natural recovery of vascular plant communities on the industrial barrens of the Sudbury Area. In: Gunn, J.M. (Ed.), Restoration and recovery of an industrial region. Progress in restoring the smelter-damaged landscape near Sud bury, Canada, pp. 93-102. Springer-Verlag, New York, Berlin, Heidelberg. Wollenweber, E., Dörr, M.. Stelzer, R., Arriaga-Giner, F.J., 1992. Lipophilic phenolics from the leaves of Empetrum nigrum - Chemical structures and exudate localization. Botanica Acta 105, 300-305. Wu. L.. 1990. Colonization and establishment of plants in contaminated environments. In Shaw, A.J. (Ed.), Heavy metal tolerance in plants: Evolutionary aspects, pp. 269-284. CRC Press, FL. USA. Yang. X.E.. Baligar, V.C.. Foster, J.C.. Martens, D.C., 1997. Accumulation and transport of nickel in relation to organic acids in ryegrass and maize grown with different nickel levels. Plant and Soil 196, 271-276. Zackrisson. 0., Nilsson. M.-C., 1992. Allelopathic effects by Empetrum hermaphroditum on seed germination of two boreal tree species. Canadian Journal of Forest Research 22, 1310-1319. Zackrisson. 0., Nilsson. M.-C.. Dahlberg, A., Jäderlund, A., 1997. Interference mechanisms in conifer-Ericaceae-feathermoss communities. Oikos 78, 209-220. I Reprinted from Environmental Pollution, 109, Monni, S., Salemaa, M., Millar, N. The tolerance of Empetrum nigrum to copper and nickel, 221-229. © 2000 with permission from Elsevier Science. 0269-7491/00, S - see front matter © 2000 Elsevier Science Ltd. All rights reserved. PII: 50269-7491(99)00264-X Environmental Pollution 109 (2000) 221-229 The tolerance of Empetrum nigrum to copper and nickel S. Monnia' b '*, M. Salemaa\ N. Miliar 0 a Vantaa Research Centre, Finnish Forest Research Institute, PO Box 18. FIS-01301 Vantaa. Finland b Department of Ecology and Systematics. University of Helsinki. FIN-00014 Helsinki. Finland c 52 Upper Queens Road, Ashford, Kent TN24 BHF, UK Received 28 May 1999; accepted 14 September 1999 "Capsule": Increasing copper and nickel concentration increased adverse effects on growth of Empetrum nigrum Abstract The Cu and Ni tolerance of 3- to 5-year-old cuttings of crowberry (Empetrum nigrum) were tested in controlled conditions. Six levels of Cu (0.1-100 mg l_l), five levels of Ni (0-100 mg 1~') and nine levels of Cu +Ni were applied. The elongation of the shoots, new shoot and root dry weights indicated an adverse effect of increasing Cu and Ni concentrations. At low Cu levels the addition of Ni decreased the dry weights more than at high Cu levels. The results show that E. nigrum accumulated high concentrations of Cu and Ni mainly in old stem tissue, which contained a maximum of over 3000 mg kg -1 Cu and 1000 mg kg -1 Ni. The concentrations of Cu and Ni in E. nigrum were higher than those measured in plants growing in areas near to Cu-Ni smelters, but the accumula tion pattern was similar. The survival of the cuttings was not affected suggesting that E. nigrum possesses an internal heavy metal tolerance. © 2000 Elsevier Science Ltd. All rights reserved. Keywords: Empetrum nigrum (crowberry); Copper; Nickel; Heavy metal tolerance; Cu and Ni accumulation 1. Introduction Recent plant ecological studies have concentrated on crowberry (Empetrum nigrum L.) because of its wide ecological amplitude (Elvebakk and Spjelkavik, 1995) and its special ecology in boreal forest ecosystems (Zackrisson et al., 1997; Nilsson et ai., 1998). E. nigrum is an evergreen dwarf shrub that occurs on a variety of substrates. It can tolerate soil pH values ranging from 2.5 to at least 7.7 (Bell and Tallis, 1973). In the northern hemisphere E. nigrum is limited to the cooler regions, but it is also common in mountains at lower latitudes (Good, 1927; Bell and Tallis, 1973). In boreal forests it grows on nutrient-poor, light dry heaths (Sarvas, 1937) and can form single-species plant communities in clear cut areas (Nilsson, 1992). It also grows on ombrogenous peat and peaty podsols (Good, 1927; Bell and Tallis, 1973). One interesting feature of E. nigrum is its ability to influence the establishment of other species; its leaf extracts inhibit, for example, the establishment of Scots * Corresponding author. Tel.: +358-9-857051; fax: + 358-9- 85705569 E-mail address: satu.monni@metla.fi (S. Monni). pine (Pinus sylvestris L.) seedlings (e.g. Nilsson and Zackrisson, 1992; Zackrisson and Nilsson, 1992; Nils son, 1994). E. nigrum is one of the few understorey species which grows on severely heavy-metal contaminated sites in the vicinity of copper-nickel (Cu-Ni) smelters in the north ern hemisphere (Laaksovirta and Silvola, 1975; Chertov et al., 1993; Helmisaari et al., 1995; Uhlig et al., 1996; Shevtsova, 1998). For instance, it survives at a distance of 0.5 km from the Cu-Ni smelter at Harjavalta, southwest (SW) Finland, where almost all the other plant species have disappeared. There have been severe changes in the environment at this site (Helmisaari et al., 1995); the total Cu concentration in the organic layer is over 5800 mg kg -1 d.m., Ni concentration 460 mg kg -1 , while base cations (e.g. Ca 2 + , Mg 2 + ) have been displaced from the topsoil. The concentrations of other heavy metals (Fe, Zn, Cd, Pb, Cr) in the organic layer are also high near the smelter (Derome and Lin droos, 1998). An ability to grow in heavy-metal polluted areas makes E. nigrum a suitable species for the reme diation of contaminated soil. It is one of the plant species which has been used in a revegetation experiment in the vicinity of the Harjavalta smelter (Kiikkilä et al., 1996). 222 S. Monni et al. j Environmental Pollution 109 (2000) 221-229 In general, plants possess several mechanisms en abling them to tolerate high heavy metal concentrations (e.g. Antonovics et al., 1971; Baker, 1981, 1987; Wool house, 1983). The metals may accumulate in older tissue or in senescing leaves that are shed (Reilly, 1969). The ericoid mycorrhiza of dwarf shrubs has been suggested to accumulate heavy metals in the roots leading to a reduction in heavy metal concentrations in the shoot (Bradley et al., 1981, 1952; Burt, 1954). Even if the rate of heavy metal uptake can be controlled by plants, total avoidance of metal uptake is not possible (e.g. Baker and Walker, 1990). Heavy metals are mainly taken up via the roots, but uptake via leaves and stems has also been discussed (Chamberlain, 1983; Koricheva et al., 1997). The heavy metal tolerance mechanism of E. nigrum is not well known. Besides the accumulation of metals in older plant parts and the restriction of their transport to the younger parts (Helmisaari et al., 1995; Uhlig et al., 1996), the xeromorphic structure of both the leaves and stems (Miller, 1975; Carlquist, 1989; Wollenweber et al, 1992) might also help survival in heavy-metal polluted areas (Pandolfini et al., 1996), where the water-holding capacity of the soil is usually decreased (Derome and Nieminen, 1998). Brooks et al. (1977, 1980) defined a plant to be a hyperaccumulator of Cu and Ni if the tis sue concentration exceeds 1000 mg kg -1 . Based on field studies (e.g. Chertov et al., 1993; Uhlig et al., 1996), it can be suggested that E. nigrum is a hyperaccumulator of Cu and Ni. However, the elevated Cu and Ni levels measured in field samples may be partly due to surface contamination. It is impossible to differentiate between the toxic effects of different heavy metals on a plant in field con ditions. It is also difficult to assess aerial deposition of the metals on the surfaces of leaves and stems compared to the inner accumulation of metals taken up by the roots. In the present study the Cu and Ni tolerance of E. nigrum is studied for the first time under controlled conditions. The aim of the experiment was: (1) to investigate the growth and leaf discolouration of E. nigrum cuttings exposed to elevated levels of Cu, Ni or to both metals: (2) to determine the interactive effects of these metals; (3) to determine the accumulation pattern of Cu and Ni in different plant parts; and (4) to com pare the responses in the experiment to those observed in the field. 2. Materials and methods 2.1. Experimental setup The 3- to 5-year-old cuttings of E. nigrum used in the experiment originated from an unpolluted area in SW Finland. They were rooted in a mixture of sand and peat. When the experiment was being set up the roots of the cuttings were washed, and 15 root samples were taken to check mycorrhizal infection by a modification of the method of Koske and Gemma (1989). All the fine roots examined were moderately colonized by ericoid mycorrhiza. The roots were cut back to 12 cm to stand ardize the experiment material and to give the roots sufficient growing space. The roots were dipped in acti vated carbon to be able to separate, at the end of experiment, stained old roots from white new roots formed during the study. For the experiment the plants were randomised and transferred to plastic pots con taining quartz sand (0.5 1). The cuttings were watered twice a week (50 ml/pot) with a nutrient solution modified by Striblev and Read (1976) containing the following nutrients (mg l" 1 ): P. 11.3 (KH : PO,); K, 28.5 (KC1); Ca. 29.2 (CaCl2x2H 2o); Mg. 8.8 (MgCl 2 x6H 20); Fe, 3 (FeChxöH.O); Mn, 0.5 (MnCU); B, 0.5 (H 3B0 3); Zn, 0.1 (ZnCl 2 ); Mo. 0.1 (Na2Moo4); N, 65 (ammonium and nitrate as NH4NO3 and ammonium as NH 4CI in the ratio NH 4:N0 3 70:30). Six levels of Cu (0.1, 1, 10, 22, 46 and 100 mg 1~'), five levels of Ni (0, 10, 22, 46 and 100 mg l -1 ) and nine combinations of Cu and Ni [Cu (mg 1~')/Ni (mg l" 1 ): 10/10, 22/10, 46/10, 100/10, 22/22, 46/22, 100/22, 46/46, 100/46] were used. Cu was given as CuS0 4 and Ni as NiCli. The control solution in the Ni series con tained 0.1 mg l _l Cu, because Cu is a micronutrient. In the Cu series the control solution contained 1 mg l" 1 Cu. The Cu and Ni concentrations used were selected according to preliminary studies. The concentrations measured in the organic layer near the Harjavalta Cu- Ni smelter represent the real ecological conditions for E. nigrum in this polluted environment. Because the Cu concentrations in the soil at Harjavalta are higher than the Ni concentrations (Derome and Lindroos, 1998), we studied how the Cu response changed when a less or equal concentration of Ni was combined with the Cu treatment. This explains why all the combinations were not applied and experiment was not fully factorial. The experimental design was completely random, and the total number of seedlings treated was 152 (eight plants/ treatment). The experiment was carried out in the greenhouse of the Finnish Forest Research Institute at Ruotsinkylä (60°21' N, 25°00' E) over the period 1 July to 13 August 1996. The light conditions were natural, and the mean temperature was regulated at +2O°C during the day and + 15° C at night. The relative humidity in the greenhouse was approximately 60-70%. 2.2. Estimating the influence of heavy metals on E nigrum The variables used to assess metal toxicity were elongation of the main stem and two side branches S. Monni et ai./ Environmental Pollution 109 (2000J 221-229 223 and the leaf, stem, and root dry weights. The accumu lation and translocation of Cu and Ni by the roots to aboveground parts of the plant were determined by chemical analysis from different plant parts. The total aboveground uptake of Cu and Ni was determined by the following equation: total Cu or Ni uptake (mg) = total aboveground biomass (kg) x total Cu or Ni concentration (mg kg -1 dry wt.) in the aboveground parts. The total initial length of the cuttings was measured before planting, and the new growth during the experi ment calculated as the difference between the initial and final lengths of the leading shoot and side branches. For biomass analyses the material was divided into the following ll parts: (1) green leaves of current-year growth; (2) discoloured leaves of current-year growth; (3) green leaves of previous-year growth; (4) dis coloured leaves of previous-year grow th; (5) older green leaves; (6) older discoloured leaves; (7) stems of current-year growth; (8) stems of previous-year growth; (9) older stems; (10) new roots; and (11) old roots. The samples were homogenized, dried at +6O°C, and weighed, and the total Cu and Ni concentrations of plant parts 1 to 9 were determined by dry ashing ( + 550° C), extraction of the ash with 0.2 M HCI, and analysis on an inductively coupled plasma atomic emission spectrometer (ICP-AES) (Dahlquist and Knoll, 1978). Eight replicate samples were combined to obtain enough material for the chemical analyses. The current and previous-year leaves and current and previous-year stems were also combined. In addition, Cu and Ni concentrations were analysed similarly from the green leaves and stems of the cuttings not treated with metals, to know the initial Cu and Ni concentra tions of these cuttings. 2.3. Statistical analysis Two-factor analysis of variance (ANOVA) was used in analysing the effects of Cu and Ni and their interactions on the length growth and biomass vari ables of E. nigrum (GLM procedure, SAS Institute Inc., 1994; Sokal and Rohlf, 1995). A few missing values were replaced by the group mean. Pairwise comparisons between the treatments and between the Cu and Ni series were performed by the r-test. The initial length of the cuttings was used as the covariate in the variance model. Logarithmic trans formations lg (.Y+l) were made to normalise the data. Relationships between Cu and Ni concentrations in plant parts and applied concentrations were studied by regression models (REG and NLIN procedures. SAS Inst.). Regression equations are also given for Cu and Ni uptake as a function of applied Cu and Ni amounts (SAS Institute Inc., 1994; Sokal and Rohlf, 1995). 3. Results 3.1. The effect of Cu and Ni on shoot elongation Elevated Cu or Ni concentrations in the nutrient solution decreased the elongation of the leading shoot and side branches (Fig. la, b), and the effects were sig nificant (p < 0.01) (Table 2). The overall survival was not affected and only four plants died during the following treatments: Cu 10 mg M (one plant), Ni 100 mg 1"' (two plants), Cu 22 mg l _l , Ni 10 mg l - ' (one plant). The length growth was the highest at a concentration of 1 mg I_l1 _1 Cu for the leading shoot and at 0.1 mg l _l Cu for the side branches. The leading shoot and the two side branches behaved differently, with elongation of the lead ing shoot generally being better than elongation of the side branches when treated with Cu (Fig. la). Ni inhibited the elongation of the leading shoot more than Cu (/-test Fig. 1. The mean length growth (mm) of the leading shoot and side branches of Empetrum nigrum treated with a nutrient solution con taining different concentrations (mg 1-1) of (a) copper (Cu) or (b) nickel (Ni). The bar indicates the standard error (n = 8). Means which do not differ significantly between treatments (p > 0.05) are marked with similar capital letters for the leading shoot and small letters for the side branches. 224 S. Monni et ai. / Environmental Pollution 109 (2000) 221-229 between Cu and Ni series, p< 0.01). Growth retardation started at lower Ni concentrations than at comparable Cu concentrations (Fig. la, b). Compared to the highest growth (1 mg I_l1 _1 Cu, omg 1"' Ni for leading shoot and 0.1 mg l" 1 Cu, omg l - ' Ni for side branches), the growth reduction of the leading shoot was 79% and of the side branches 72% when treated with 100 mg I -1 Cu. The corresponding reductions for Ni were 96 and 85%. In general, the addition of Ni limited growth to a similar extent at all Cu levels, because the interaction term was not significant (Tables l and 2). However, when Ni was added to the highest Cu level (100 mg l - ') there were no differences in length growth compared to the plants treated with only Cu (100 mg 1""') (Fig. la, Table 1). 3.2. The effect of Cu and Ni on biomass The dry weight of the current-year shoots (leaves and stems combined) decreased significantly (/?< 0.001) Table 1 Means for length growth (mm) and biomass (mg) of Empetrum nigrum treated with combinations of Cu and Nia a /i =B. except for side branches // = 6-8. Significant differences between the treatment means (elongation of leading shoots and dry weight or discoloured leaves) are marked with different letters. Other wise there were no significant differences between the treatment means by /-test (/? < 0.05). with increasing Cu or Ni concentrations in the nutrient solution (Table 2). The dry weight was the highest with the 0.1 mg I~' Cu and 0 mg l -1 Ni treatment (Fig. 2a). Cu suppressed biomass production more than Ni (t-test between Cu and Ni series, p < 0.05). Compared to the highest dry weight (treatment 0.1 mg l" 1 Cu, 0 mg l _l Fig 2. The mean dry biomass (mg) of the current-year growth of Empetrum nigrum treated with a nutrient solution containing different concentrations of (a) nickel (Ni) or copper (Cu) (mg 1-1) or (b) com binations of both Ni and Cu. The bar indicates the standard error (n = 8). Means which do not differ significantly between treatments (p>0.05) are marked with similar capital letters for the Ni treatment and small letters for the Cu treatment. Table 2 F and p values of two-factor analysis of variance (ANOVA) for log-transformed response variables a a Initial length of leading shoot has been used as a covariate. ns. not significant, Cu and Ni Elongation Dry weight applied (mg l" 1) Cu/Ni Leading Side New- New Discoloured shoots branches shoots roots leaves (mm) (mm) (mg) (mg) (mg) 10/10 13 a 7 42 2 14 a 22/10 8 ab 6 36 3 27 ab 46/10 4 b 4 33 3 28 ab 100/10 5ab 3 29 2 34 ab 22/22 6 ab 3 23 1 1 8 ab 46/22 5 b 7 36 1 52b 100/22 3 b 5 31 1 33 ab 46/46 4 b 3 23 1 33 ab 100 46 4 b 3 27 3 27ab Source of variation df Elongation Dry weight Leading shoot Side branches New shoots New roots Discoloured leaves F P F P F P F P F P Cu 5 4.73 0.001 3.24 0.009 10.25 0.000 5.32 0.000 1.76 ns Ni 4 7.14 0.000 7.05 0.000 6.12 0.000 3.41 0.011 5.41 0.001 Cu.Ni 9 0.61 ns 1.20 ns 4.11 0.000 2.67 0.007 1.44 ns Length 1 0.50 ns 0.00 ns 11.88 0.001 0.00 ns 43.57 0.000 S. Monni et ai. / Environmental Pollution 109 (2000) 221-229 225 Ni), the dry weight of the current-year shoot was limited by 78% for plants treated with 100 mg 1"' Cu and 79% when treated with 100 mg 1"' Ni. The statistical interaction between Cu and Ni in the dry weights was significant (/> < 0.001) (Table 2). The effect of adding Ni varied at different Cu levels. At low Cu levels (Cu <46 mg l _l ), Ni increased the growth suppression, but at higher Cu concentrations (46, 100 mg l" 1) there was no effect (Fig. 2b). Compared to the highest dry weight, Cu and Ni in combination sup pressed the dry weight by a maximum of 86% when 22 mg l _l Cu and 22 mg l -1 Ni or 46 mg I -1 Cu and 46 mg l" 1 Ni were applied. The dry weight of the discoloured leaves increased as the Ni concentration increased and the effect was sig nificant (p = 0.001) (Table 2). A similar trend was also observed with increasing Cu concentrations (Fig. 3a), Fig. 3. The mean dry biomass (mg) of discoloured leaves of the pre vious and current-year growth of Empetrum nigrum treated with a nutrient solution containing different concentrations of (a) nickel (Ni) or copper (Cu) (mg 1"') or (b) both Ni and Cu in combination. The bar indicates the standard error (n = 8). Means which do not differ significantly between treatments (p > 0.05) are marked with similar capital letters for the Ni treatment and small letters for the Cu treatment. but the effect was not significant (Table 2). At the lowest Cu and Ni concentrations, 1-4% of the total leaf bio mass was discoloured in the current and previous-year growths. The maximum proportion of discoloured leaves out of the total leaf biomass was 43% with the highest Cu concentration and 56% with the highest Ni concentration. Apart from the highest Cu concentration (100 mg l -1), adding Ni to the Cu solution increased discoloura tion (Fig. 3b). If 10 or 22 mg I_l1 _1 Ni were added to the 22 mg l _l Cu solution, or 22 mg 1"' Ni were added to the 46 mg I~' Cu solution, the discolouration increased significantly (/><0.05) compared to the treatments in which corresponding amounts of Cu only were applied. The highest maximum proportion of discoloured leaves was 66% when 46 mg I -1 Cu and 46 mg 1"' Ni were given in combination. The amount of new root biomass produced was less if Cu or Ni was increased (/» < 0.05) (Fig. 4, Table 2). The maximum proportion of new roots out of the total root biomass was 25% when treated with 1 mg I -1 Cu and 0 mg 1"' Ni. The growth of new roots was inhibited more at lower Cu and Ni levels compared to the growth decrease of the aboveground biomass. The root growth was only about 1-4% of the total below ground biomass when the Cu or Ni concentrations in the nutrient solution were more than 22 mg l - '. The statistical interaction term between Cu and Ni was significant (/> < 0.01) on root growth (Table 2). Ni decreased root growth at lower Cu levels, but the response varied at the highest concentrations (Fig. 4, Table I). The proportion of new root dry weight out of the total below ground biomass was at a maximum of 6% when 46 mg l~' Cu and 10 mg 1"' Ni were given. Fig. 4. The mean new root biomass (mg) of Empetrum nigrum treated with a nutrient solution containing Cu or Ni (mg l-1 ). The bar indicates the standard error (n = 8). Means which do not differ significantly between treatments (p > 0.05) are marked with similar capital letters for the Ni treatment and small letters for the Cu treatment. 226 S. Monni et ai. / Environmental Pollution 109 (2000) 221-229 3.3. Cli and Ni accumulation in E. nigrum Not treated cuttings contained Ni below detection limit and Cu less than 13 and 21 mg kg - ' in green leaves and stems, respectively. During the 6-week course of the experiment E. nigrum accumulated increasing amounts of Cu and Ni with increasing Cu or Ni levels in the nutrient solution (Table 3). Metal accumulation increased with age, with younger parts containing less Cu and Ni than older ones. Discoloured leaves contained more Cu and Ni than the green leaves of corresponding age classes. The highest concentration of Cu in the stem tissue was 2033 mg kg -1 and of Ni 2132 mg kg -1 of plants treated with 100 mg l - ' of Cu or Ni alone. The total uptake of Cu and Ni increased linearly as Cu and Ni concentrations in the solution increased (Table 4). The proportion of Cu and Ni taken up out of the amount applied [lOO x total Cu or Ni uptake (mg)/added Cu or Ni (mg)] decreased as the solution concentrations increased, being at its maximum about 2% (Table 4). If treated with Cu and Ni in combination, E. nigrum accumulated very high concentrations of both metals in the living parts. Older stems contained over 2700 mg kg -1 Cu and about 2600 mg kg -1 Ni if treated with 46 mg l -1 Cu and 46 mg l - ' Ni together. If the Cu con centration in the nutrient solution was 100 mg l - '-and the Ni concentration 46 mg l - ', older stems contained 3038 mg kg - ' Cu and 1267 mg kg"' Ni (Table 3). The accumulation pattern was similar to that if only Cu or Ni was applied. The concentrations of Cu and Ni in different plant parts varied depending on whether Cu and Ni were given in combination or separately. However, the sum of the Cu and Ni concentrations were generally higher when given separately than in combination in leaves (Table 3). In older stems the sum of the Cu and Ni concentrations applied separately were higher than con centrations of these metals given in combination in the following treatments: Cu/Ni 10/10, 22/10, 46/10. As the amount of Cu and Ni applied increased, the sum Table 3 Copper (Cu) and nickel (Ni) concentrations (mg kg -1 dry wt.) in young (current and previous-year growths) and older parts (older than current and previous-year growths) of Empctrum nigrum treated with different concentrations of Cu and Ni (one bulk sample comprises eight plants)a 3 <, Below detection limit; -, not enough sample material for the analyses or a temporary fault in the ashing oven. Cu concentration in older green leaves is given in parentheses. Regression equations for concentration in plant part (v) as a function of applied concentration (.t) are given for data sets (five samples) marked with letters b-g. b .1 = 26.43+ 1.63a—0.01.v 2 , r 2 = 0.98, /?< 0.05. c v = 50.0 + 157.73.V 0 56 , r 2 = 0.98. p < 0.05. d > =9.81 + 1.60.x, r 1 = 0.9\,p<0.05. e .v=12.18 + 9.30-v 0 JO . r = 0.81, ns. ' y = 30.0 + 36.46.V 0 88, r 2 = 0.99, p < 0.01. S v = 4.73+ 0.60.Y 1 49 , r 2 = 0.99, Applied conc. Green leaves, young Discoloured leaves, older Discoloured leaves, young Stems, older Stems, young (mg l" 1) Cu/Ni Cu Ni Cu Ni Cu Ni Cu Ni Cu Ni 1/0 26 (30) b - 16 76 e 22 d 10/0 52 - 53 514 44 22/0 42 465 - 875 28 46/0 78 265 98 1755 58 100/0 52 - 141 2033 183 0.1/0 15 e _ _ 31 f 7» 0.1/10 34 - < 252 40 0.1/22 17 313 90 522 32 0.1/46 97 - 335 1217 203 0.1/100 56 - 192 2132 583 10/10 28 22 _ . _ 24 < 180 167 _ _ 22/10 - - - - - - 603 259 57 32 46/10 36 11 353 205 51 < 1272 247 - - 100/10 40 < 327 196 124 14 2567 237 60 14 22/22 _ _ 318 374 52 24 _ - 90 102 46/22 29 25 310 385 82 32 - - - - 100/22 33 < - - 690 131 - - - - 46/46 26 < 484 613 83 62 2705 2594 32 136 100/46 28 16 - - 54 20 3038 1267 30 < 5. Monni et al. / Environmental Pollution 109 (2000) 221-229 227 Table 4 Total copper (Cu) and nickel (Ni) uptake (mg) during the experiment measured in the total aboveground biomass a a This is based on the equation: the total aboveground biomass (kg)x total Cu or Ni concentration in the aboveground parts of Empetrum nigrum (mg kg -1 ). The Cu and Ni concentrations in the aboveground parts used in the calculations (see Table 3). Regression equations for the total Cu or Ni uptake in aboveground parts (r) as a function of applied Cu or Ni (.v) are given for data sets (five samples) marked with the letters b and c. b ; = 0.029 + 0.003.V, r 2 = 0.92, p< 0.05. c y = 0.006 + 0.004.V, r 2 = 1,p<0.001. of the Cu and Ni concentrations in stems were lower if these metals were applied separately than if they were applied in combination (100/10,46/46, 100/46) (Table 3). 3.4. Discussion The growth inhibition of E. nigrum with increasing Cu and Ni concentrations in the nutrient solution indi cated that Cu and Ni had toxic effects on plants. The clearest responses to Cu and Ni were in the dry weights of the shoots and roots, as has also been reported earlier for other plant species (e.g. Roth et al., 1971; Taylor and Crowder, 1983; Kramer et al., 1996; L'Huillier et al., 1996). Root growth was affected at relatively low levels of these metals, and it decreased already by more than half of the lowest treatments (0.1 or Img l - ' Cu and 0 mg 1"' Ni) compared to the treatment of 10 mg I~' Cu or Ni. In field conditions .the growth of E. nigrum has also clearly decreased in Cu- and Ni-polluted areas compared to the populations growing in a clean back ground area (Shevtsova, 1998). The growth suppression is the cost of tolerance and, although growth was affec ted, the survival did not decrease during the short experimental period. The 100% survival in high Cu and Ni treatments would indicate that populations originat ing from clean areas can also tolerate these high metal concentrations. However, survival at the highest Cu and Ni concentrations would have probably been affected if the experiment had continued. E. nigrum tolerated high accumulated concentrations of Cu and Ni, with the highest concentrations being in the old stems. The Cu and Ni concentrations in dis coloured leaves were higher than those in green leaves of the same age. The accumulation pattern of these metals within the plant was in good agreement with the values reported for E. nigrum growing in the field (Helmisaari et al., 1995; Uhlig et al., 1996). The current annual liv ing shoots of E. nigrum have been reported to contain 180 mg kg"' Cu, the older living parts 1450 mg kg - ', and dead biomass 4130 mg kg - ' at a distance of 0.5 km from the Harjavalta smelter (Helmisaari et al, 1995). In aboveground parts, Ni, Fe, Pb and Zn have also been reported to accumulate with age, the highest concentra tions occurring in old stems. Relatively lower con centrations of Cu are found in the roots compared to the aboveground parts (Uhlig et al., 1996). The accu mulation pattern was similar throughout the whole concentration range. Therefore, high concentrations of metals in the older stem would indicate root-to-shoot transport. The results suggest that E. nigrum possesses an internal tolerance mechanism, which is supported by greenhouse and previous field studies (Uhlig et al., 1996). Kramer et al. (1996) suggested that root-to-shoot transport is important in Ni tolerance and hyper accumulation. E. nigrum accumulated over 2000 mg kg -1 Cu or Ni in the living tissue treated with 100 mg I~' Cu or Ni alone and more than 2700 mg kg -1 Cu and about 2600 mg kg -1 Ni if treated simultaneously with 46 mg l~ l Cu and 46 mg l _l Ni. These concentrations were higher than those measured earlier in the living tissue of E. nigrum, the highest reported Cu concentration in older living parts being 1500 mg kg -1 (Helmisaari et al., 1995) and the Ni concentration in leaves 1510 mg kg"' (Chertov et al., 1993). The experiment indicates that the high concentrations found in the field only partly can be due to surface contamination. The concentrations meas ured in living parts in the field and greenhouse are higher than the limit suggested for hyperaccumulation (1000 mg kg -1 ) (Brooks et al., 1977, 1980). An other ericaceous dwarf shrub, Vaccinium uliginosum L., has also been found to accumulate as high Cu concentra tions in the field (DiLabio and Rencz, 1980). The effect of adding Ni to the Cu solution suppressed the dry weights of the roots and shoots slightly more than Cu alone. However, the combined application of Cu and Ni suppressed the dry weights less than the summed effects of Cu and Ni alone. Also the sum of the Cu and Ni concentrations in the plant parts were generally higher in the plants treated with Cu and Ni separately than those treated with Cu and Ni in com bination. Only in plants treated with the highest Applied Cu (mg) Total Cu uptake (mg) b Uptake (%) Applied Ni (mg) Total Ni uptake (mg) c Uptake (%) 0.6 0.0117 1.95 0 0.0059 _ 6 0.0528 0.S8 6 0.0284 0.47 13.2 0.0982 0.74 13.2 0.0575 0.44 27.6 0.0964 0.35 27.6 0.0989 0.36 60 0.2093 0.35 60 0.2225 0.37 228 S. Monni et al.j Environmental Pollution 109 (2000) 221-229 concentrations of Cu and Ni, were the sum of these metals in the older stems higher when applied in com bination than separately. The plant concentrations indicate that, at lower concentrations, there is competi tive interaction between the uptake of these two metals. When the concentration of a toxic metal increases, adding another toxic metal does not further limit growth. However, interactions between metals are often complex, and they are dependent on the metal con centration and pH in the growth medium (Balsberg- Pählsson, 1989). The concentrations of Cu in the plant parts treated only with the control Cu levels (0.1 or 1 mg 1~') were high compared to background values measured in the field (Helmisaari et ai., 1995; Uhlig et ai., 1996). Field conditions are different from those in the greenhouse where, for instance, organic matter and soil organisms are missing. Earlier manipulation studies have shown that roots growing in a medium with only low con centrations of metals are able to take up large amounts of these metals into the plant. Medappa and Dana (1970) showed that cranberry (Vaccinium macroccirpon Ait.) contained 600 mg A 1 kg -1 in the roots when the external solution concentration was only 2.5 mg l - '. The roots of Salix cciprea L. exposed to trace con centrations of Fe in a nutrient solution contained 1680 mg kg -1 Fe root dry wt. (Talbot and Etherington, 1987). Taylor and Crowder (1983) reported that Typha latifolia L. accumulated higher Cu and Ni concentra tions in the greenhouse than in the field than would have been expected on the basis of the Cu and Ni con centrations in the growth substrate. The use of activated carbon to determine the biomass of roots grown during this experiment may have influ enced the experimental conditions. Activated carbon has a detoxifying effect on toxic compounds (Shaw et al., 1990; Zackrisson and Nilsson, 1992), and charcoal have increased the colonization of ericoid mycorrhizas in Vaccinium sp. (Duclos and Fortin, 1983). The exact plant-available concentrations of the elements are not known because activated carbon has most probably immobilized some of the Cu and Ni. 4. Conclusions This greenhouse experiment showed that E. nigrum tolerated high Cu and Ni concentrations in its living parts, and helps to explain occurrence of this plant in highly polluted areas near the smelters. Although growth was suppressed, survival was not affected by the metals. The greenhouse studies showed that Cu and Ni concentrations were the highest in the stems indicating root-to-shoot transport. Transport to the green leaves appears to have been restricted, as reflected by the rela tively constant Cu and Ni concentrations in the green leaves. Therefore, it is suggested that E. nigrum pos sesses an internal Cu and Ni tolerance. Acknowledgements We thank the staff of the Finnish Forest Research Institute for helping with the practical work and at the Ruotsinkylä greenhouse for taking good care of the plants. We also thank Maarit Martikainen for per forming the chemical analyses and Erkki Tomppo for helping with statistical analyses. We would like to extend special thanks to Christian Uhlig for valuable discussions throughout the study and for commenting on the manuscript. We also thank Heljä-Sisko Helmi saari and Tiina Nieminen for making comments on the manuscript and John Derome for revising the language. The work was funded by the Nature Conservation Fund of Finland and the Jenny and. Antti Wihuri Foundation. References Antonovics, J., Bradshaw, A.D., Turner, R.G., 1971. Heavy metal tolerance in plants. Advanced Ecological Research 7, 1-85. Baker, A.J.M., 1981. Accumulators and excluders—strategies in the response of plants to heavy metals. Journal of Plant Nutrition 3, 643-654. Baker, A.J.M., 1987. Metal tolerance. New Phytologist 106, 93-111 Baker, A.J.M., Walker, P.L., 1990. Ecophysiology of metal uptake by tolerant plants. In: Shaw, A.J. (Ed.), Heavy Metal Tolerance in Plants: Evolutionary Aspects. CRC Press, Boca Raton, FL, pp. 155-178. Balsberg-Pählsson, A.-M., 1989. Toxicity of heavy metals (Zn, Cu, Cd, Pb) to vascular plants. A literature review. Water, Air and Soil Pollution 47, 287-319. Bell, J.N.8., Tallis, J.H., 1973. Biological flora of the British Isles. Empetrum nigrum L. Journal of Ecology 61, 289-305. Bradley, R., Burt, A.J., Read, D.J., 1981. Mycorrhizal infection and resistance to heavy metal toxicity in Calluna vulgaris. Nature 292, 335-337. Bradley, R., Burt, A.J., Read, D.J., 1982. The biology of mycorrhiza in the Ericaceae. VIII. The role of mycorrhizal infection in heavy metal resistance. New Phytologist 91, 197-209. Brooks, R.R., Lee, J., Reeves, R.D., Jaffre, T., 1977. Detection of nickeliferous rocks by analysis of herbarium specimens of indicator plants. Journal of Geochemical Exploration 7, 49-57. Brooks, R.R., Reeves, R.D., Morrison, R.S., Malaisse, F., 1980. Hyperaccumulation of copper and cobalt—a review. Bulletin de la Societe Royale de Botanique de Belgique 113, 166-172. Burt, A.J., 1984. Ecophysiological studies of heavy metal resistance in Calluna vulgaris (L.) Hull, with special reference to mycorrhizal infection. Doctoral thesis, University of Sheffield, UK. Carlquist, S., 1989. Wood and bark anatomy of Empetraceae; com ments on paedomorphosis in woods of certain small shrubs. Aliso 12, 497-515. Chamberlain, A.C., 1983. Fallout of lead and uptake by crops, Atmospheric Environment 17, 693-706. Chertov, 0.G., Nadporozshskaya. M.A., Lapshina, 1.N., Grigorieva, 0.A., 1993. Soil degradation in surroundings of "Pechenganikel" smelter complex. In: Kozlov, M.V., Haukioja, E.. Yarmishko. V.T. (Eds.), Aerial Pollution in Kola Peninsula. Proceedings of the International Workshop. 14-16 April, St. Petersburg. Kola Scien tific Centre. Apatity. Russia, pp. 159-162. S. Monni et ai. I Environmental Pollution 109 (2000) 221-229 229 Dahlquist, R.L., Knoll, J.W., 1978. Inductively coupled plasma atomic emission spectrometry: analysis of biological materials and soils for major, trace, and ultra-trace elements. Applied Spectro scopy 32, 1-30. Derome, J., Lindroos, A.J., 1998. Effects of heavy metal contamina tion on macronutrient availability and acidification parameters in forest soil in the vicinity of the Harjavalta Cu-Ni smelter, SW Fin land. Environmental Pollution 99, 225-232. Derome, J., Nieminen, T.. 1998. Metal and macronutrient fluxes in heavy-metal polluted Scots pine ecosystems in SW Finland. Envi ronmental Pollution 103, 219-228. DiLabio, R.N.W.. Rencz. A.N., 1980. Relationship between levels of copper, uranium, and lead in glacial sediments and in Vaccinium uliginosum at an arctic site enriched with heavy metals. Canadian Journal of Botany 58, 2017-2021. Duclos, J.L., Fortin, J.A., 1983. Effect of glucose and active charcoal on in vitro synthesis of ericoid mycorrhiza with Vaccinium spp. New Phytologist 94, 95-102. Elvebakk, A.. Spjelkavik, S., 1995. The ecology and distribution of Empetrum nigrum ssp. hermaphroditum on Svalbard and Jan Mayen. Nordic Journal of Botany 15, 541-552. Good. R.D'O., 1927. The genus Empetrum L. Journal of the Linnean Society of London 47, 489-523. Helmisaari, H.-S., Derome, J., Fritze. H., Nieminen, T., Palmgren, K., Salemaa, M., Vanha-Majamaa, 1., 1995. Copper in Scots pine for ests around a heavy-metal smelter in South-Western Finland. Water. Air and Soil Pollution 85, 1727-1732. Kiikkilä. 0., Uhlig, C., Helmisaari, H.-S., Salemaa, M„ Fritze, H., Nieminen, T., Vanha-Majamaa, 1., Derome, J., 1996. Revegeta tion experiment in heavy metal contaminated Scots pine forest at Harjavalta, western Finland. In: Kopponen P., Kärenlampi, S., Rekilä, R., Kärenlampi, L. (Eds.), Bio- and Ecotechnological Methods in Restoration. Abstracts of an International Advanced Course and Minisymposium, 16-18 December. Kuopio, Finland. Kuopio University, p. 9. Koricheva, J., Roy, S., Vranjic, J.A., Haukioja, E., Hughes, P.R., Hänninen, 0., 1997. Antioxidant responses to simulated acid rain and heavy metal deposition in birch seedlings. Environmental Pol lution 95, 249-258. Koske, R.E., Gemma, J.N., 1989. A modified procedure for stain ing roots to detect VA mycorrhizas. Mycological Research 92, 486- 505. Krämer, U., Cotter-Howells, J.D., Charnock, J.M., Baker, A.J.M., Smith, J.A.C., 1996. Free histidine as a metal chelator in plants that accumulate nickel. Nature 379, 635-638. Laaksovirta, K., Silvola, J., 1975. Effect of air pollution by copper, sulphuric acid and fertilizer factories on plants at Harjavalta, W Finland. Annales Botanici Fennici 12, 81-88. L'Huillier, L., d'Auzac, J., Durand, M., Michaud-Ferriere, N., 1996. Nickel effects on two maize (Zea mays) cultivars: growth, structure, Ni concentration, and localization. Canadian Journal of Botany 74, 1547-1554. Medappa, K.C., Dana, M.N., 1970. Tolerance of cranberry plants to manganese, iron and aluminium. Journal of the American Society for Horticultural Science 95, 107-110. Miller. H.J., 1975. Anatomical characteristics of some woody plants of the Angmagssalik district of southeast Greenland. Meddelelser om Gronland 198. 1-30. Nilsson, M.-C.. 1992. The mechanisms of biological interference by Empetrum hermaphroditum on tree seedling establishment in boreal forest ecosystems. Doctoral thesis. Swedish University of Agri cultural Sciences. Umea. Sweden. Nilsson, M.-C., 1994. Separation of allelopathy and resource compe tition by the boreal dwarf shrub Empetrum hermaphroditum Hagerup. Oecologia 9S, 1-7. Nilsson, M.-C., Zackrisson, 0., 1992. Inhibition of Scots pine seedling establishment by Empetrum hermaphroditum. Journal of Chemical Ecology 18, 1857-1870. Nilsson, M.-C., Gallet. C., Wallstedt, A., 1998. Temporal variability of phenolics and batatasin-111 Empetrum hermaphroditum leaves over an eight-year period: interpretations of ecological function. Oikos 81, 6-16. Pandolfini, T., Gabbrielli, R., Ciscato, M., 1996. Nickel toxicity in two Durum wheat cultivars differing in drought sensitivity. Journal of Plant Nutrition 19, 1611-1627. Reilly. C., 1969. The uptake and accumulation of copper by Becium homblei (De Wild.) Duvign. & Plance. New Phytologist 68, 1081— 1087. Roth, J.A., Wallihan. E.F., Sharpless, R.G., 1971. Uptake by oats and soybeans of copper and nickel added to a peat soil. Soil Science 112. 338-342. Sarvas, R., 1937. Havaintoja kasvillisuuden kehityksestä Pohjois suomen kuloaloilla. Silva Fennica 44, 1-64. SAS Institute Inc., 1994. SAS/STAT User's Guide. Version 6. 4th Edition. SAS Institute Inc., Cary. NC. Shaw, G., Leake, J.R., Baker, A.J.M., Read, D.J., 1990. The biology of mycorrhiza in the Ericaceae. XVII. The role of mycorrhizal infection in the regulation of iron uptake by ericaceous plants. New Phytologist 115, 251-258. Shevtsova, A. 1998. Responses of subarctic dwarf shrubs to climate change and air pollution. Annales Universitatis Turkuensis Ser. All, 113. Doctoral thesis. Sokal, R.R., Rohlf, F.J., 1995. Biometry. W.H. Freeman, New York Stribley, D.P., Read, D.J., 1976. The biology of mycorrhiza in the Ericaceae. VI. The effects of mycorrhizal infection and concentra tion of ammonium nitrogen on growth of cranberry (Vaccinium microcarpon Ait.) in sand culture. New Phytologist 77, 63-72. Talbot, R.J., Etherington, J.R., 1987. Comparative studies of plant growth and distribution in relation to waterlogging XIII. The effect of Fe 2+ on photosynthesis and respiration of Salix caprea and S. cinerea ssp. olei/olia. New Phytologist 105, 575-583. Taylor, G.J., Crowder, A.A., 1983. Uptake and accumulation of cop per, nickel, and iron by Typha latifolia grown in solution culture. Canadian Journal of Botany 61, 1825-1830. Uhlig, C., Salemaa, M., Vanha-Majamaa, 1., 1996. Element distribu tion in Empetrum nigrum microsites at heavy metal contaminated sites in Harjavalta, Western Finland. In: Kopponen P., Kärenlampi, S., "Rekilä, R., Kärenlampi, L. (Eds.), Bio- and Ecotechnological Methods in Restoration. Abstracts of an International Advanced Course and Minisymposium, 16-18 December. Kuopio, Finland, p. 21. Wollenweber, E., Dörr, M., Stelzer, R., Arriaga-Giner, F.J., 1992. Lipophilic phenolics from the leaves of Empetrum nigrum—chemical structures and exudate localization. Botanica Acta 105, 300-305. Woolhouse, H.W., 1983. Toxicity and tolerance in the responses of plants to metals. In: Lange, 0.L., Nobel. P.S., Osmond. C.8., Zieg ler. H. (Eds.), Encyclopedia of Plant Physiology, Vol. 12. Springer. Berlin, pp. 245-300. Zackrisson. 0., Nilsson, M.-C.. 1992. Allelopathic effects by Empe trum hermaphroditum on seed germination of two boreal tree spe cies. Canadian Journal of Forest Research 22, 1310-1319. Zackrisson. 0.. Nilsson. M.-C., Dahlberg. A.. Jäderlund, A., 1997. Interference mechanisms in conifer-Ericaceae-feathermoss commu nities. Oikos 78. 209-220. II Reprinted from Environmental Pollution, 109, Monni, S., Salemaa, M., White, C., Tuittila, E., Huopalainen, M. Copper resistance of Calluna vulgaris originating from the pollution gradient of a Cu-Ni smelter, in southwest Finland, 211-219. © 2000 with permission from Elsevier Science. 0269-7491 00; S - see front matter ( 2000 Elsevier Science Ltd. All rights reserved. PII; 50269-7491 (99)00265-1 Environmental Pollution 109 (2000) 211-219 Copper resistance of Calluna vulgaris originating from the pollution gradient of a Cu-Ni smelter, in southwest Finland S. Monni a' b'*, M. Salemaa 3 , C. White c , E. Tuittila d , M. Huopalainen d a Vantaa Research Centre, Finnish Forest Research Institute. PO Box 18. FIS-01301, Vantaa, Finland b Department of Ecology and Systematics, University of Helsinki, PO Box 7, FIN-00014 Helsinki. Finland c Higrove. Nedstop. Oldcroft, Lydney, Gloucestershire GLI 5 4i\'Q, UK d Department of Forest Ecology. University of Helsinki, PO Box 24. FIN-00014 Helsinki. Finland Received 28 May 1999; accepted 14 September 1999 "Capsule": Heather seedlings from seed collected at three locations did not differ appreciably in their measured responses to copper treatments. Abstract The copper (Cu) resistance of 1-year-old seedlings of heather (Callwta vulgaris) was tested in a greenhouse experiment. The plant material originated from seeds collected from three peatland sites located 1.2 km to the NW, and 2.5 and 5.5 km to the NE of the Harjavalta Cu-nickel (Ni) smelter, SW Finland. The plants were watered with a nutrient solution containing five different levels of Cu (1, 10, 22, 46 and 100 mg l - '). Cu clearly decreased the length growth of shoots, shoot and root biomass of C. vulgaris. More than 50% of the seedlings exposed to the highest Cu treatment died. C. vulgaris accumulated high amounts of Cu, the living old roots containing a maximum of 2200 mg kg" 1 Cu and the living stems 1300 mg kg -1 Cu. Discolouring leaves contained higher Cu concentrations than green leaves. The results indicate Cu accumulation in roots and root-to-shoot transport. Some differences were found between the responses of the three seed provenances, but none of the populations proved to be more resistant to Cu than the others in all the measured responses. © 2000 Elsevier Science Ltd. All rights reserved. Keywords: Calluna vulgaris (heather); Copper; Heavy-metal resistance; Cu accumulation; Cu-Ni smelter 1. Introduction Heather (Calluna vulgaris (L.) Hull.) is a widespread and common species in Europe that usually grows on acidic and nutrient-poor soils (Gimingham, 1960, 1972). It is found on heathlands (e.g. Mohamed and Giming ham, 1970; Haapasaari, 1988), moors, bogs, fixed sand dunes and in forests (Gimingham, 1960). In Finland, C. vulgaris is among the 10 most common vascular plant species in forest and peatland vegetation (National Forest Inventory, 1995, unpublished results). The litter of C. vulgaris is rich in phenolic compounds (Jalal and Read, 1983), which decompose slowly and modify the soil causing acidification and organic matter accumula tion. This leads to soil conditions which are optimal for * Corresponding author. Tel.: +358-9-857051: fax: +358-9-8570 5569 E-mail address: satu.monni(a metla.fi (S. Monni). the growth of C. vulgaris and impairs the growing conditions of other species (Grubb and Suter, 1971; Robinson, 1971; Haslam, 1977; Leake, 1988). The dominance of C. vulgaris in nutrient-poor environments has been explained on the basis of the effective use of nutrients; the ericoid mycorrhizal endophyte of C. vul garis roots is also able to utilise organic sources of nitrogen and carbon (Bajwa et al., 1985; Leake and Read, 1989). Resistance to heavy metals can be achieved by two strategies: tolerance and avoidance (Baker, 1987). Tol erant populations of many plant species have been found in heavy-metal-polluted mine and smelter areas (e.g. Schat and Ten Bookum, 1992). Tolerance to heavy metals may be either based on the evolution of tolerant genotypes (ecotypes) or may be environmentally induced (phenotypic plasticy) (e.g. Antonovics et al., 1971; Baker, 1987). There are indications that C. vul garis possesses constitutive tolerance, and ecotypic 212 S. Monni et oi. / Environmental Pollution 109 (2000) 211-219 differentiation of C. vulgaris populations growing in polluted or non-polluted areas has been suggested (Burt, 1984). In the middle of its distribution area, e.g. in Great Britain and Germany, C. vulgaris has been found to be very tolerant to heavy metals occurring on a variety of substrates, e.g. serpentine and polluted soils (Marrs and Bannister, 1978; Eltrop et al.. 1991). C. vul garis is one of the species colonizing waste heaps in metal-polluted areas (Burt, 1984). In the northern parts of its distribution, however, C. vulgaris is absent from severely heavy-metal-polluted areas in Fennoscandia (Gilbert. 1975; Laaksovirta and Silvola, 1975; Väisänen, 1986; Salemaa and Vanha-Majamaa, 1993) and the Kola Peninsula (Mikhail Kozlov, unpublished results), and it is considered to be a sensitive species to metals (Laaksovirta and Silvola, 1975). In general, plants possess physiological mechanisms that enable them to resist high substrate heavy-metal concentrations (Antonovics et al., 1971; Baker, 1981, 1987; Woolhouse, 1953). Plants can either detoxify metals by binding them with organic acids, proteins, etc. (Lee et al., 1978; Rauser and Curvetto, 1980), or accu mulate the metals in the different plant parts or cell organelles (Reilly, 1969; Bringezu, et al., 1999). Copper (Cu) is essential for plant growth and is involved in many metabolic processes, e.g. photosynthesis. How ever, high concentrations are toxic to plants (Marsch ner, 1995). Cu may accumulate in older plant parts (Helmisaari et al., 1995), such as stems (Uhlig et al., 1996) or senescing leaves (Reilly, 1969; Ernst, 1972). Bradley et al. (1981, 1982) and Burt (1984) showed that the ericoid mycorrhiza of C. vulgaris accumulates elevated concentrations of Cu and prevents transloca tion of metals to the growing parts of the plant. Even if the rate of heavy metal uptake can be controlled by plants, total avoidance of metal uptake is not possible (e.g. Baker and Walker, 1990). The ecology (Gimingham, 1972) and heavy-metal resistance (Bradley et al., 1981, 1982; Burt, 1984) of C. vulgaris have been extensively studied in oceanic heath lands, but its response to heavy metals in boreal forests and peatlands is not known. The aim of this study was to investigate: (1) the Cu resistance of northern popula tions of C. vulgaris under controlled conditions; (2) whether the populations originating from sites closest to a copper-nickel (Cu-Ni) smelter are the most resistant ones; and (3) the pattern of Cu accumulation within different parts of C. vulgaris. 2. Materials and methods 2.1. Seed material of C. vulgaris The seeds of C. vulgaris originated from peatland seedbanks at three sites, located at different distances from the Cu-Ni smelter at Harjavalta, southwest (SW) Finland. The Cu smelter was established in 1945 and the Ni smelter in 1960. and sulphur dioxide and heavy metals have been emitted into the environment for the past 40-50 years. The deposition of metals near the smelter was considerably reduced in the 1990s after a new, taller stack and electrostatic filters were built (Rantalahti. 1995). The prevailing wind direction is from the SW and the emissions are primarily dispersed to the northeast (NE). The seedbank samples were collected from three peatland sites located 1.2 km to the northwest (NW) (Lammaistensuo) and 2.5 km (Kotosuo) and 5.5 km to the NE (Pyhäsuo) of the smelter. Metal concentrations in the surface peat decreased with increasing distance from the pollution source; at 0.9 km the total Cu con centration was 3560 mg kg" 1 dry wt. and total Ni concentration 470 mg kg~' dry wt., and at 2.4 km 1870 mg kg~' dry wt. Cu and 340 mg kg~' dry wt. Ni. The corresponding concentrations at a distance of 5.0 km were 770 mg kg" 1 Cu and 180 mg kg -1 Ni. Metals were extracted with acid (dry digestion) by inductively cou pled plasma atomic emission spectrometry (ICP-AES). Soil pH was 3.5-3.7 (Veijalainen, 1998). The abundance of C. vulgaris in the understorey vegetation increases with increasing distance from the pollution source. It appears for the first time at 2.3 km with a point frequency of 0.6%. At 2.8 km it is 2.4% and at 5.7 km 11.3% (Antti Reinikainen, unpublished results). However, results from seedbank experiments show that viable seeds of C. vulgaris are found at 1.2 km, which do not germinate in the field, but germinate in optimal light and moisture conditions, in greenhouse. Seedling mortality increases with decreasing distance from the pollution source (Salemaa and Uotila. 1996; Huopalainen, 1998). The seedbank samples (0-5 cm soil layer) of C. vul garis were collected on 27-29 September 1995 using a sxlo-cm corer, using methods described by Huopalai nen (1998). The samples were stored in the dark at + 5°C for 5 weeks (chilling). After chilling, the samples were transferred to a greenhouse where they were spread evenly in a 1-cm-thick layer on a mixture of horticultural peat and quartz sand in trays to maintain the optimal light conditions (Wesson and Wareing. 1969). When the seeds had germinated and seedlings grown sufficiently, they were planted in a mixture of peat and quartz sand and left to grow until the experi ment started. 2.2. Plant culture system Nine-month-old seedlings of C. vulgaris were used in the experiment. The roots of the seedlings were washed with tap water and cut back to a length of 12 cm in order to standardize the experiment material and to give 5. Monni el ai. / Environmental Pollution 109 (2000) 211-219 213 the roots sufficient growing space. The roots were dip ped in activated carbon so as to facilitate separation of the stained old roots from white new ones at the end of the experiment. The plants were transferred to plastic pots containing quartz sand (0.5 1) and randomised for different Cu treatments. The cuttings were watered twice weekly (50 ml/pot) with a nutrient solution modified by Stribley and Read (1976) containing the following nutrients (mg 1~'): P, 11.3 (KHiPOj); K, 28.5 (KCl)fCa, 29.2 (CaCk2H 2 O): Mg. 8.8 (MgCl2.6H,0); Fe, 3 (FeCI 3 .6H : O); Mn. 0.5 (MnCli); B, 0.5 (H,B03 ); Zn, 0.1 (ZnCU); Mo, 0.1 (Na 2Moo 4); N, 65 (ammonium and nitrate as NH4NO3 and ammonium as NH 4 CI in the ratio NH 4 :N0 3 , 70:30). Five concentrations of Cu (1, 10, 22, 46 and 100 mg 1~'), applied as CuS0 4, were used with 10 replicates per treatment. The first treatment (1 mg l" 1 ) represented the control. The total number of seed lings was 150. The experiment was carried out in the greenhouse of the Finnish Forest Research Institute at Ruotsinkylä (60°21' N, 25°00' E) over the period 12 June to 25 July 1996, under natural photoperiods (light period approximately 18 h, dark period 6 h). The mean tem perature was regulated to +2O°C during the day and + 15° C at night, and the relative humidity above 60%. 2.3. The response variables of C. vulgaris The parameters used to assess metal toxicity were elongation of the leading shoot and two side branches, and the biomass (dry wt.) of the leaves, stems and roots. The root and shoot uptake of Cu and the allocation of different elements to the leaves, stems and roots were also determined. The total initial length of the leading shoot and two side branches were measured before planting, and the growth during the experiment determined as the dif ference between the initial and final lengths of the stems of the leading shoots and side branches. The material was divided into six parts for biomass analyses: (1) green leaves; (2) discoloured leaves; (3) stems of leading shoots; (4) stems of side branches; (5) new roots; and (6) old roots. The samples were homogenized, dried at + 60° C, weighed and dry digested (+ 550 3 C). The extraction of the ash was done with 2-3 ml 6 M HCI (pro-analys) in water bath (approximately + 80° C). The dry residue was diluted with 10 ml 1 M HCI for 20 min and filtered (filter paper; Schleicher & Schuell 589-') with 0.1 M HCI. The final volume of the solution was 25-50 ml depend ing on the sample weight. The total element (Cu, Ni, P, K, Mg, Ca, Fe) concentrations of the plant parts were determined by inductively coupled plasma atomic emis sion spectrometry (ICP-AES). Ten replicate samples were combined in order to obtain sufficient material for the chemical analyses. 2.4. Statistical analysis Two-factor analysis of variance (ANOVA) was used to analyse the effects of Cu treatment, the origin of the seedlings and their interaction with the response vari ables of C. vulgaris (GLM procedure, SAS Institute Inc.. 1994). The initial length of the leading shoot was used as the covariate in the variance model. Logarith mic transformations lg (,r + 1) were performed in order to normalise the data. Pairwise comparisons between the treatments were performed by the /-test. Pearson correlations were calculated between the concentrations of Cu and other elements in the different plant parts. The model-predicted length growth of the leading shoot and two side branches, shoot and leaf biomass are shown in Figs. 1-3, in which the initial length of the leading shoot is used as a covariate. Fig. 1. The mean length growth (mm) of (a) leading shoot and (b) side branches of Calluna vulgaris predicted by the variance model in which the initial length of the seedlings was used as the covariate (n=10/Cu treatment). Seedlings were collected at three distances from the smelter and treated with a nutrient solution containing different concentra tions (mg l-1 ) of Cu. 214 5. Monni et aI.J Environmental Pollution 109 f2OOOJ 211-219 Fig. 2. The mean aboveground biomass (mg) of Callitna vulgaris pre dicted by the variance model in which the initial length of the seedlings was used as the covariate (n = 10/Cu treatment). See details in Fig. 1. Fig. 3. The mean biomass (mg) of (a) green and (b) discoloured leaves of Calluna vulgaris predicted by the variance model in which the initial length of the seedlings was used as the covariate (n = 10 Cu treatment). See details in Fig. 1. 3. Results 3.1. The influence of Cu oil length grow th and biomass Increasing Cu concentrations caused a clear decrease in the length growth of leading shoot and side branches of C. vulgaris (Fig. la, b). The highest Cu treatment almost completely inhibited growth (Fig. 1), the max imum growth reduction of the leading shoots being 86- 99% and that of the side branches 84-92% compared to that for the control treatment (1 mg l" 1 Cu). The growth of the leading shoot was much higher than that of the side branches at low concentrations, but not at the two highest Cu concentrations (Table I). The mean length growth of leading shoot and side branches dif fered significantly (/>< 0.001) between the Cu treat ments (Table 2). The origin of the seedlings also had a statistically significant effect on the growth of the lead ing shoots (p< 0.001) (Table 2). The model-predicted length growth of the seedlings originating from 2.5 km of the smelter was the highest in almost all of the Cu treatments (Fig. la, b). The shoot biomass decreased as the Cu concentration in the solution increased, the difference between the treatments being significant (p< 0.001) (Table 2). The decrease was significant when the Cu concentration increased from Itolo mg l - ' in one origin, but the decreasing trend was not clear at higher concentrations (Table 1). The shoot biomass decreased less than the length growth, the maximal reduction in biomass vary ing from 52 to 67% in the different seedling origins. The effect of origin was significant (p<0.001) (Table 2). The model-predicted shoot biomass values of the seed lings from a distance of 2.5 km were again generally the highest (Fig. 2). The green leaf biomass decreased and leaf discoloura tion increased clearly as the Cu concentrations in the nutrient solution increased (Table 1). The effect of the Cu treatments on both leaf biomasses was significant (p< 0.001) (Table 2). The proportion of discoloured leaves out of the total leaf biomass was 5-8% at the lowest Cu concentrations, but 74-85% at the highest Cu concentration. The mean green leaf biomasses differed significantly between the origins (p < 0.05) (Table 2). The model-predicted biomasses of the green and discoloured leaves were generally lower in the 1.2 km origin, but no trends were found in the other origins (Fig. 3a, b). The statistical interaction term between the origin and the Cu concentration in green leaf biomass was significant (p< 0.01) (Table 2). Increasing Cu concentrations caused a stronger decrease in the new root biomass than in shoot biomass (Figs. 2 and 4). However, shoot/root ratio stayed rela tively constant throughout the whole concentration range (Table 1). The mean root biomass was significantly different between the Cu treatments (p < 0.001) (Table 2). S. Monni et ai. / Environmental Pollution 109 (2000J 211-219 215 Table 1 The means (x) and standard errors (SE) of growth (mm) and biomass (mg), shoot/root (S/R) ratio and survival (%) of Calluna vulgaris treated with different concentrations of copper (Cu) a a The means which did not differ significantly between treatments (p>0.05) are marked with similar letters. Table 2 F and p values of two-factor analysis of variance (ANOVA) for the log-transformed response variables Fig. 4. The mean new root biomass (mg) of Calluna vulgaris treated with a nutrient solution containing different concentrations of Cu (mg l-1) (n = 10, Cu treatment). See details in Fig. 1. Root growth was almost totally inhibited by 100 mg l -1 Cu, the proportion of new roots being only 2-3% at the highest Cu concentrations. 3.2. Survival of C. vulgaris seedlings Cu clearly caused a decrease in the survival of the C. vulgaris seedlings. The average mortality was 10, 20 and 60% for Cu concentrations of 22, 46 and 100 mg l -1 . The mortality of the seedlings varied between the origins in the different treatments, but none of the ori gins had the highest survival rate in all the treatments (Table 1). 3.3. Cu accumulation in C. vulgaris Cu concentrations increased in all parts of C. vulgaris with increasing Cu levels in the nutrient solution (Fig. 5). Source of variation df Elongation Dry weight Leading shoots Side branches Aboveground Green leaves Discoloured leaves New roots F P F P F P F P F P F P Cu 4 38.36 0.0001 34.68 0.0001 11.14 0.0001 30.04 0.0001 32.42 0.0001 31.53 0.0001 Origin 2 7.56 0.0008 1.55 0.2171 12.37 0.0001 3.50 0.0330 2.92 0.0572 1.36 0.2606 Cu-origin 8 1.3 0.2491 0.08 0.9996 0.99 0.4455 2.81 0.0065 1.00 0.4410 0.81 0.5983 Length 1 8.43 0.0043 0.93 0.3368 88.51 0.0001 27.30 0.0001 18.03 0.000 1 7.85 0.0058 Cu applied Elongation Dry weight Survival (mg 1" -') (%) Cu Origin Leading shoot Side branches Aboveground Green leaves Discoloured leaves New roots S/R (km) (mm) (mm) (mg) (mg) (mg) (mg) ratio lv SE X' SE X SE X SE X SE X SE 1 1.2 49 J 6.8 22 a 4.5 383 abc 43 237 acd 28 21 ab 4.8 69 a 21.1 2.7 100 1 2.5 41 a 6.9 21 a 3.9 463 ad 97 306 a 61 16b 4.1 39 a 7.1 2.6 100 1 5.5 44 a 8.0 21 a 4.6 516 ab 95 340 a 63 28 abe 5.8 38 ab 10.0 2.7 100 10 1.2 25 a 6.3 13 ae 3.6 201 cf 37 120 acd 24 14 ab 2.4 20 ab 4.2 3.1 100 10 2.5 25 ah 7.5 13 a 2.0 218 bed 37 137 ad 26 15 ab 4.7 14 ab 3.4 2.0 100 10 5.5 15 bh 6.0 11 ab 2.3 253 ce 47 153 ab 34 26 abc 6.0 14 bc 3.5 2.1 100 22 1.2 11 bef 2.3 6 cdf 2.0 212 efh 22 89 bd 9 41 ac 9.8 5 de 1.7 3.0 100 22 2.5 18 aeh 3.7 7 bee 1.6 237 be 60 136 acd 38 32 cef 6.6 11 bcf 3.4 2.2 80 22 5.5 11 bf 3.5 5 cd 1.5 167 egf 30 75 be 18 38 cfti 6.0 3 de 1.2 3.0 90 46 1.2 8 cdf 3.9 3 cdf 0.7 168 gh 29 46 e 18 59 df 7.7 2 de 0.9 3.0 50 46 2.5 7 bdg 2.2 5 cdf 2.1 267 be 53 110 bd 33 81 di 14.8 9 ce 3.8 2.4 90 46 5.5 3 c 1.3 2 dg 0.5 307 c 56 140 ab 36 72 dgh 12.2 11 def 7.0 2.4 100 100 1.2 1 c 0.3 2 df 0.6 125 gh 18 28 e 9 52 df 6.5 2 de 1.6 3.3 50 100 2.5 6 bd 1.4 3 cdf 1.1 224 ce 35 17 e 7 138 gi 30.2 2 d 1.3 2.4 40 100 5.5 3 eg 1.4 2 fg 0.9 196 egf 37 28 e 12 106 di 23.8 1 d 0.7 2.8 40 216 S. Monni el ai. I Environmental Pollution 109 (2000) 211-219 Fig. 5. The mean Cu concentrations (mg kg -1) in the leaves, stems and roots of Calluna vulgaris from the three collection sites. Plants were exposed to different Cu concentrations in a nutrient solution (10 plants in one composite sample). The individual plant parts had varying Cu concentra tions and the concentrations decreased generally in the following order: old roots > new roots > stems > discoloured leaves > green leaves (Fig. 5). The Cu con centrations were extremely high in old roots and stems, exceeding 5500 and 4500 mg kg" 1 , respectively. Because the dead and living plants were not analysed separately, these results also include dead parts. The highest Cu concentrations in the living parts of old roots exceeded 2200 mg kg" 1 and in the stems of the leading shoots 1300 mg kg" 1 . The lowest concentrations were found in the green leaves (maximum 300 mg kg" 1), while the discoloured leaves contained up to 1600 mg kg" 1 Cu (Fig. 5). The proportion of Cu taken up from the nutrient solution (lOOxroot and shoot uptake/added Cu) was the highest at the lowest Cu concentrations and it was about 8% in the lowest Cu treatment (Table 3). It decreased with increasing Cu concentration in the nutrient solution, and the relative shoot and root uptake was only about 2% in the concentration of 10 mg l" 1 Cu in nutrient solution. Relative root uptake decreased almost as much as relative shoot uptake with increasing Cu concentration in the nutrient solution. Also approximately the amount of Cu in the shoots was as high as that in the roots (Table 3). 3.4. Correlations between Cu and other elements in the plant parts of C. vulgaris There were interactions between Cu and the other elements. The uptake of P, Ca, K and Mg decreased as the Cu concentration increased in the stems of the leading shoots and side branches indicating that Cu interfered with nutrient uptake. The uptake of K increased and that of Ca and Fe decreased as the Cu concentration increased in the discoloured leaves of C. vulgaris (Table 4). S. Monni et ai. / Environmental Pollution 109 (2000) 211-219 217 Table 3 Shoot and root Cu uptake (µg) and relative uptake of Cu (percentage of the uptaken amount of Cu applied) of Calluna vulgaris originating from three distances from the smelter Table 4 The Pearson correlations between copper (Cu) concentrations (mg kg -1 ) and the other elements in different plant parts of Calluna vulgaris a a // = 15, equals the number of Cu concentrations applied to three origins (5x3). "p < 0.05. 4. Discussion Increasing Cu concentrations in the nutrient solution had clearly an adverse effect on the growth parameters of C. vulgaris compared to plants treated with the con trol solution (I mg I -1 Cu). Burt (1984) found that in the treatment of 80 mg I~' Cu, the root growth of C. vulgaris was completely inhibited, while shoot biomass increased during the experiment. Also in this study, the root biomass was more strongly inhibited than the shoot biomass, indicating the toxic effects of Cu especially on the roots. C. vulgaris appeared to accumulate high concentra tions of Cu because the surviving roots and stems con tained over 1000 mg kg -1 Cu. The concentrations were at the same level as in the study of Burt (1984) on non mycorrhizal plants. Bradley et al. (1981, 1982) reported even higher shoot and root concentrations in mycor rhizal and non-mycorrhizal plants. Although C. vulgaris is capable of accumulating high concentrations of Cu in greenhouse conditions within a relatively short period, the Cu concentrations of plants growing in Cu-polluted soil have been considerably lower. In Great Britain, the Cu concentration in C. vulgaris fine roots was only about 140 mg kg -1 and in brown shoots 90 mg kg -1 , even though the total Cu concentration in the spoil heap was 2500 mg kg - ' (Burt, 1984). 115 mg kg -1 Cu has been reported in green shoots (Marrs and Bannister, 1978). The greenhouse experiment clearly showed that northern populations of C. vulgaris also resisted high Cu concentrations. However, because the duration of the experiment was relatively short and mortality increased strongly at the highest Cu concentrations, it is probable that long-term exposure is detrimental to C. vulgaris seedlings. It has been demonstrated that, although the seeds are capable of germination, seedling survival is short in seedbank soil sampled from polluted areas (Salemaa and Uotila, 1996; Huopalainen, 1998). Not only the origin of C. vulgaris, but also the environ mental conditions may regulate the occurrence of C. vulgaris and explain why it grows in heavy-metal pol luted areas on oceanic heathlands but not in boreal forests. In addition to high metal concentrations (Derome and Lindroos, 1998; Veijalainen, 1998), also drainage and deficiency of base cations near the Cu-Ni Applied Cu Origin distance from Root uptake Shoot uptake Total uptake Relative root Relative shoot Relative total (10 3 ng) the smelter (km) (Mg) (Mg) (Mg) uptake (%) uptake (%) uptake (%) 0.6 1.2 21 18 39 3.5 3.0 6.5 0.6 2.5 30 22 52 5.0 3.7 8.7 0.6 5.5 24 21 45 4.0 3.5 7.5 6.0 1.2 38 17 55 0.6 0.3 0.9 6.0 2.5 81 22 103 1.3 0.4 1.7 6.0 5.5 124 35 159 2.1 0.6 2.7 13.2 1.2 156 99 255 1.2 0.7 1.9 13.2 2.5 173 57 230 1.3 0.4 1.7 13.2 5.5 107 58 165 0.8 0.4 1.2 27.6 1.2 203 130 333 0.7 0.5 1.2 27.6 2.5 289 225 514 1.0 0.8 1.9 27.6 5.5 277 132 409 1.0 0.5 1.5 60.0 1.2 209 238 446 0.3 0.4 0.7 60.0 2.5 452 435 887 0.8 0.7 1.5 60.0 5.5 332 321 653 0.6 0.5 1.1 Elements in the different parts of the plant P Ca K Mg Fe Cu in the stems of leading shoots -0.86- -0.75* -0.76* -0.74* 0.03 Cu in the stems of side branches -0.58* -0.55' -0.58* -0.61* 0.27 Cu in the green leaves 0.20 0.50 0.34 0.40 -0.10 Cu in the discoloured leaves 0.33 -0.70* 0.65* -0.19 -0.58* 218 5. Monni et ai. I Environmental Pollution 109 (2000) 211-219 smelter at Harjavalta (Derome and Nieminen, 1995) may explain the absence of C. vulgaris. The water sup ply is known to be critical for seedling establishment of C. vulgaris (e.g. Gimingham, 1972). It has been suggested that the Cu resistance of C. vulgaris is based on an exclusion mechanism, in which the ericoid mycorrhiza of C. vulgaris roots accumulates the metals and thus prevents Cu transportation to the shoots (Bradley et al., 1981, 1982; Burt, 1984). Indica tions of true Cu tolerance in C. vulgaris have also been reported (Burt. 1954). According to our data the role of mycorrhiza in heavy metal resistance cannot be eval uated. However, the tolerance of C. vulgaris growing in the polluted soil may develop before infection of the fungus in the roots (Burt. 1984), thus suggesting that the plant itself also tolerates high metal concentra tions. The higher Cu concentrations found in the roots than in the shoots support the hypothesis of Bradley et al. (1981, 1982). However, based on the high Cu con centrations in the stems, our results indicate root-to shoot transport suggesting that C. vulgaris possesses true tolerance to Cu. The discoloured leaves had higher Cu concentrations than the green leaves, which indi cated that C. vulgaris also transports Cu to the leaves. This is supported by Marrs and Bannister (1978) and Burt (1984), who concluded that C. vulgaris could detoxify and remove metals by accumulating them in older shedding leaves and stems. In the beginning of the experiment, activated carbon was used to stain the old roots as it is a good method for distinguishing between old and new roots for the biomass measurements. However, activated carbon is known to absorb metals (Shaw et al., 1990), and this might have altered the experimental conditions. For instance, the Cu concentrations in the old roots might have been overestimated to some extent, even though the roots were washed before the chemical analyses. Some Cu may also have remained on the root surface. In con trast, the shoot Cu concentrations more certainly reflect the Cu taken up by the roots from the nutrient solution. Compared to another evergreen dwarf shrub, Empe trum nigrum, which commonly grows close to smelters (Helmisaari et al., 1995), the growth of C. vulgaris decreased to a relatively greater extent under compar able concentrations. When exposed to a concentration of 100 mg I_l1 _1 Cu in the nutrient solution, the survival of E. nigrum was not affected even though the Cu con centrations in the living stem tissue were more than 3000 mg kg" 1 (Monni et al., 1999). Because the survival of C. vulgaris was strongly affected at the highest con centrations, it appears to be more sensitive to Cu than E. nigrum. The different age of the seedlings may, how ever, affect the result. In contrast to the findings of this study, Burt (1984) suggested ecotypic differences in Cu tolerance between C. vulgaris seedlings originating from metal-contaminated mine spoil and a non-polluted background area. Sig nificant differences between the origin of the seedlings were only found in certain growth parameters, but no systematic tendency was seen. Neither did Cu accumu lation give any indications of differences between the origins. Because the most distant origin of the seedlings was 5.5 km from the smelter and not from a non polluted background area, this might be the reason why no clear ecotypic differences were found. In addition, there are suggestions that the evolution of tolerant races of woody plants does not occur as frequently as for grasses and herbs (Antonovics et al., 1971). 5. Conclusions The exposure of C. vulgaris seedlings to Cu caused clear toxicity symptoms in the roots of the plants, and survival was strongly affected. The shoot biomass also decreased, though it was not so clearly affected as root growth. However, C. vulgaris accumulated high con centrations of Cu in the living parts during the short exposure period, and the accumulation pattern was similar throughout the concentration range. We can conclude that C. vulgaris tolerated Cu by accumulating it especially in the roots and stems, as well as in the discolouring leaves. There were some indications of ecotypic differences between the origins but the respon ses were not related to the distance from the smelter. Acknowledgements We thank the staff of the Finnish Forest Research Institute for helping with the practical work and at the Ruotsinkylä greenhouse for taking good care of the plants. We thank Ilkka Vanha-Majamaa for helping to produce the seedling material for the study and Maarit Martikainen for performing the chemical analyses. We would like to express special thanks to Christian Uhlig for the valuable discussions and critical comments on the manuscript, and to John Derome for revising the language. The work was funded by the Suomen Biolo gian Seura Vanamo and the Jenny and Antti Wihuri Foundation. References Antonovics, J., Bradshaw, A.D., Turner, R.G.. 1971. Heavy metal tolerance in plants. Advances in Ecological Research 7, 1-85. Bajwa, R., Abuarghub. S., Read, D.J.. 1985. The biology of mycor rhiza in the Ericaceae. X. The utilization of proteins and the pro duction of proteolytic enzymes by the mycorrhizal endophyte and by mycorrhizal plants. New Phytologist 101, 469—486. Baker, A.J.M., 1981. Accumulators and excluders—strategies in the response of plants to heavy metals. Journal of Plant Nutrition 3. 643-654. 5. Monni et ai. / Environmental Pollution 109 (2000) 211-219 219 Baker. A.J.M., 1987. Metal tolerance. New Phytologist 106, 93-111 Baker, A.J.M., Walker, P.L., 1990. Ecophysiology of metal uptake by tolerant plants. In: Shaw, A.J. (Ed.), Heavy Metal Tolerance in Plants: Evolutionary Aspects. CRC Press. FL, pp. 155-177. Bradley. R.. Burt. A.J., Read, D.J., 1981. Mycorrhizal infection and resistance to heavy metal toxicity in Calluna vulgaris. Nature 292, 335-337. Bradley. R.. Burt. A.J., Read, D.J., 1982. The biology of mycorrhiza in the Ericaceae. VIII. The role of mycorrhizal infection in heavy metal resistance. New Phytologist 91, 197-209. Bringezu, K.. Lichtenberger, 0., Leopold. 1., Neumann, D., 1999. Heavy metal tolerance of Silene vulgaris. Journal of Plant Physiol ogy 154. 536-546. Burt, A.J., 1984. Ecophysiological Studies of Heavy Metal Resistance in Calluna vulgaris (L.) Hull, with Special Reference to Mycorrhizal Infection. Doctoral thesis. University of Sheffield, UK. Derome. J.. Lindroos, A.J., 1998. Effects of heavy metal contamina tion on macronutrient availability and acidification parameters in forest soil in the vicinity of the Harjavalta Cu-Ni smelter. SW Fin land. Environmental Pollution 99, 225-232. Derome. J., Nieminen. T.. 1998. Metal and macronutrient fluxes in heavy-metal polluted Scots pine ecosystems in SW Finland. Envi ronmental Pollution 103, 219-228. Eltrop, L.. Brown, G.. Joachim, 0., Brinkmann, K... 1991. Lead tol erance of Betula and Salix in the mining area of Mechernich/Ger many. Plant and Soil 131, 275-285. Ernst. W., 1972. Ecophysiological studies on heavy metal plants in South Central Africa. Kirkia 8, 125-145. Gilbert. 0.L., 1975. Effects of air pollution on landscape and land-use around Norwegian aluminium smelters. Environmental Pollution 8, 113-121. Gimingham, C.H., 1960. Biological flora of the British isles. Calluna vulgaris (L.) Hull. Journal of Ecology 48, 455-453. Gimingham, C.H., 1972. Ecology of Heathlands. John Wiley, New York. Grubb. P.J., Suter, M.8., 1971. The mechanism of acidification of soil by Calluna and Ule.x and the significance for conservation. In: Duf fey, E., Watt, A.S. (Eds.), The Scientific Management of Animal and Plant Communities for Conservation. Blackwell Science, Oxford, pp. 115-133. Haapasaari, M., 1988. The oligotrophic heath vegetation of northern Fennoscandia and its zonation. Acta Botanica Fennica 135, 1-219. Haslam, E., 1977. Symmetry and promiscuity in procyanidin bio chemistry. Phytochemistry 16, 1625-1640. Helmisaari, H.-S., Derome, J., Fritze, H., Nieminen, T., Palmgren, K., Salemaa, M., Vanha-Majamaa, 1., 1995. Copper in Scots pine for ests around a heavy-metal smelter in South-Western Finland. Water. Air and Soil Pollution 85, 1727-1732. Huopalainen, M., 1998. Ihmistoiminnan muuttamien suoekosystee mien siemen-ja itiöpankki. MSc thesis, University of Helsinki, Finland. Jalal, M.A.F.. Read. D.J., 1983. The organic acid composition of Cal luna heathland soil with special reference to phyto- and fungitoxicity. 11. Monthly quantitative determination of the organic acid content of Calluna and spruce dominated soils. Plant and Soil 70, 273-286. Laaksovirta. K.. Silvola, J., 1975. Effect of air pollution by copper, sulphuric acid and fertilizer factories on plants at Harjavalta. W. Finland. Annales Botanici Fennici 12. 81-88. Leake. J.R.. 1988. Causes and Effects of Soil Acidification by Calluna vulgaris (L.) Hull, with Special Reference to the Role of Mycor rhizas. Doctoral thesis. University of Sheffield. UK. Leake. J.R.. Read. D.J., 1989. The biology of mycorrhiza in the Eri caceae. XIII. Some characteristics of the extracellular proteinase activity of the ericoid endophyte Hynienoscyphus ericae. New Phy tologist 112. 69-76. Lee. J., Reeves. R.D.. Brooks. R.R., Jaffre, T., 1978. The relation between nickel and citric acid in some nickel-accumulating plants. Phytochemistry 17, 1033-1035. Marrs, R.H., Bannister. P.. 1978. The adaptation of Callnna vul garis (L.) Hull to contrasting soil types. New Phytologist 81, 753 761. Marschner, H., 1995. Mineral Nutrition of Higher Plants, 2nd Edi tion. Academic Press, London. Mohamed, 8.F., Gimingham. C.H.. 1970. The morphology of vege tative regeneration in Calluna vulgaris. New Phytologist 69, 743- 750. Monni. S., Salemaa, M., Millar, N.. 1999. The tolerance of Empetrium nigrum to copper and nickel. Environmental Polution 109, 221-229. Rantalahti, R.. 1995. Outokummun Harjavallan tehtaiden ymp äistönsuojelusta. Vaskenviesti 3-4, 2-5. Rauser, W.E., Curvetto, N.R., 1980. Metallothionein occurs in roots of Agrostis tolerant to excess copper. Nature 287, 563-564. Reilly. C.. 1969. The uptake and accumulation of copper by Becium hoinblei(DeWi\d.) Duvig. & Plancke. New Phytologist 68,1081-1087. Robinson, R.K., 1971. Importance of soil toxicity in relation to the stability of plant communities. In: Duffey, E.. Watt. A.S. (Eds.), The Scientific Management of Animal and Plant Communities for Conservation. Blackwell Science, Oxford, pp. 105-113. Salemaa, M.. Uotila, T., 1996. Kanervan taimien hengissäsäilymisen todennäköisyys raskasmetallien saastuttamassa metsämaassa. In: Rekilä, R., Kärenlampi, L. (Eds.). sth Meeting of Finnish Plant Sciences, Voi. 45. Kuopio University Publications C, Natural and Environmental Sciences, p. 118. Salemaa. M., Vanha-Majamaa, 1.. 1993. Forest vegetation change along a pollution gradient in SW Finland. In: Tuomisto, J., Ruus kanen. J. (Eds.), First Finnish Conference of Environmental Sci ences, Vol. 14, Kuopio University Publications C, Natural and Environmental Sciences, pp. 101-104. SAS Institute Inc.. 1994. SAS/STAT User's Guide. Version 6, 4th Edition. SAS Institute Inc., Cary, NC. Schat, H., Ten Bookum, W.M., 1992. Genetic control of copper tol erance in Silene vulgaris. Heredity 68, 219-229. Shaw, G., Leake, J.R., Baker, A.J.M., Read, D.J., 1990. The biology of mycorrhiza in the Ericaceae. XVII. The role of mycorrhizal infection in the regulation of iron uptake by ericaceous plants. New Phytologist 1 15, 251-258. Stribley. D.P., Read, D.J., 1976. The biology of mycorrhiza in the Ericaceae. VI. The effects of mycorrhizal infection and concentra tion of ammonium nitrogen on growth of cranberry (Vaccinium microcarpon Ait.) in sand culture. New Phytologist 77, 63-72. Uhlig, C., Salemaa, M., Vanha-Majamaa, 1., 1996. Element distribution in Enxpetrum nigrum microsites at heavy metal contaminated sites in Harjavalta, Western Finland. In: Kopponen, P., Kärenlampi, S., Rekilä. R., Kärenlampi, L. (Eds.), Bio- and Ecotechnological Meth ods in Restoration. Abstracts of an International Advanced Course and Minisymposium. 16-18 December, Kuopio, Finland, p. 21. Väisänen, S., 1986. Effects of air pollution by metal, chemical and fertilizer plants on forest vegetation at Kokkola, W Finland. Annales Botanici Fennici 23, 305-315. Veijalainen, H.. 1998. The applicability of peat and needle analysis in heavy metal deposition surveys. Water, Air and Soil Pollution 107, 367-391. Wesson. G., Wareing. P.F., 1969. The role of light in the germination of naturally occurring population of buried weed seeds. Journal of Experimental Botany 20. 402-413. Woolhouse. H.VV., 1983. Toxicity and tolerance in the responses of plants to metals. In: Lange. 0.L.. Nobel. P.S., Osmond. C.8., Zieg ler. H. (Eds.). Encyclopedia of Plant Physiology. Springer, Berlin, pp. 245-300. Reprinted from Environmental Pollution, Monni, S., Uhlig, C., Junttila, 0., Hansen, E., Hynynen, J. Chemical composition and ecophysiological responses of Empetrum nigrum to aboveground element application, in press. © 2000 with permission from Elsevier Science. III DTD = 4.1.0 EN PO 2015p Disk used 0269-7491/00/$ - see front matter © 2000 Elsevier Science Ltd. AU rights reserved. PII: 50269-749 1 (00)00 139 -1 Environmental Pollution 0 (2000) 1-10 Chemical composition and ecophysiological responses of Empetrum nigrum to aboveground element application S. Monni a,b '*, C. Uhlig c , O. Junttila d , E. Hansen d , J. Hynynen 2 * Vantaa Research Centre, Finnish Forest Research Institute, Box 18, FIN-01301 Vantaa, Finland b Department of Ecology and Systematics, University of Helsinki, FIN-00014 Helsinki, Finland c Holt Research Centre, The Norwegian Crop Research Institute, N-9Ö05 Tromse, Norway d Department of Plant Physiology and Microbiology, University of Tromse, N-9037, Tromse, Norway Received 20 December 1999; accepted 2 May 2000 "Capsule": Heavy metals in solutions sprayed on Empetrum nigrum remained on plant surfaces and were not taken up Abstract Empetrum nigrum L. (crowberry) is one of the plants surviving near the Cu-Ni smelters in Finland and Russia. According to field observations, the roots of E. nigrum are situated below 40 cm depth and the root biomass is reduced in the polluted sites. This could cause a reduced root uptake of macronutrients and trace elements in the field and, therefore, the possible element uptake by aboveground parts of E. nigrum was studied in a greenhouse. Six different treatment solutions containing various heavy metal and macronutrient concentrations were applied to the stems and leaves of E. nigrum and the chemical composition and ecophysiological parameters were measured. Heavy metal concentrations in the leaves and stem bark, and Cu concentrations in the stems, increased with increasing metal concentrations in the spraying solutions. The bark and leaves had higher heavy metal concentrations than the stems of comparable age classes. The macronutrient and Mn concentrations in E. nigrum did not change significantly with increasing element concentrations in the spraying solution. Neither the stem water potential nor the leaf chlorophyll concentrations showed any clear response to element applications. Therefore, the element uptake by aboveground parts of E. nigrum was not confirmed by this study. However, there was a tendency to a decrease in C0 2 exchange rate and increase in foliar abscisic acid content in plants treated with the highest element concentrations. © 2000 Elsevier Science Ltd. All rights reserved. Keywords: Empetrum nigrum L.; Aboveground uptake; Chlorophyll; C02 exchange; Drought stress; Heavy metals 1. Introduction Empetrum nigrum L. is a xeromorphic (Carlquist, 1989; Wollenweber et al., 1992), evergreen dwarf shrub that occurs on a variety of substrates (Bell and Tallis, 1973). In the northern hemisphere characteristic environments are nutrient-poor, well-illuminated dry forests (Sarvas, 1937) and ombrogenous peat and peaty podsols (Good, 1927; Bell and Tallis, 1973). However, E. nigrum also occurs on serpentine soils (Proctor and Woodell, 1971). It is one of the few understorey species that grows on severely heavy metal-contaminated sites in the vicinity of Cu-Ni smelters (Laaksovirta and Silvola, 1975; Chertov et al., 1993; Helmisaari et al., • Corresponding author. Tel.: +358-9-857-051; fax: + 358-9-857- 05-569. E-mail address: satu.monni@metla.fi (S. Monni). 1995; Uhlig et al., 1996; Mälkönen et ai., 1999). In polluted areas and in a greenhouse experiment E. nigrum has been found to accumulate in excess of 1000 mg kg -1 of Cu and Ni in older stem tissue and is considered to be a species resistant to heavy metals (Uhlig et al., 1996; Monni et al., 2000 a). However, it has been shown that heavy metals can be retained on the surface of the leaves (Atteia and Dambrine, 1993) and twig axes (Wyttenbach et al., 1988) despite mechanical washing or rainfall, and this may contribute to the high concentrations observed in plants exposed to pollution. In several plant species, physiological processes such as photosynthesis and water status are sensitive to heavy metals (Lamoreux and Chaney, 1978; Rauser and Dumbroff, 1981; Becerril et al., 1989; Angelov et al., 1993; Bishnoi et al., 1993). Heavy metals have been found to inhibit electron transport in photosynthetic systems (Becerril et al., 1988) and the regenerative phase of the Calvin cycle (Weigel, 1985). In experimental 2 S. Monni et al. / Environmental Pollution O (2000) 1-10 studies, stomatal conductance and water potential of the leaves have decreased and the abacisic acid (ABA) content, which regulates the water status of the plant, increased when plants were exposed to Ni (Rauser and Dumbroff, 1981; Bishnoi et al., 1993). Plants are able to employ several strategies for survival when exposed to heavy metals (e.g. Antonovics et al., 1971; Baker, 1987). Heavy metal resistance can be based on either avoidance or tolerance mechanisms. Plants can be protected externally against metals or they can tolerate high tissue concentrations through specific physiological mechanisms (e.g. Baker, 1987). In the field, heavy metals are generally taken up via the roots, although metal uptake via the leaves and stems has also been reported (Chamberlain, 1983; Lin et al., 1995; Koricheva et al., 1997). Nothing is known about the aboveground uptake of metals or nutrients by E. nigrum. According to field observations, the roots of E. nigrum are situated below 40 cm depth and the root biomass is reduced in the polluted environments around the Cu-Ni smelters in Finland and Russia. This could cause a reduced root uptake of macronutrients and trace elements from upper soil horizons. Therefore, the possible element uptake by the aboveground parts of E. nigrum was studied by measuring (1) the element distribution and (2) the foliar chlorophyll and ABA contents, stem water potential, the dark respiration and the maximum photosynthesis of E. nigrum after heavy metal and macronutrient applications to leaves and stems. 2. Material and methods 2.1. Application of heavy metals and macronutrients The aboveground element application study was car ried out in the greenhouse of the Finnish Forest Research Institute, at Ruotsinkylä (60°21' N, 25°00' E) during 1 August-17 September 1997. The light period was natural (light period approximately 15 h, dark per iod 9 h) and the mean temperature was maintained at + 15 and +22° C during the night and day, respectively. The relative humidity in the greenhouse varied from 30 to 90%, being at its lowest in the afternoon and the highest at night. Nutrient solutions with six different element compo sitions (Table 1) were sprayed twice a week on the aerial parts of 4 to 5-year-old seedlings of E. nigrum grown in quartz sand (on the bottom) and peat substrate in pots. The composition of wet deposition near the Cu-Ni smelter at Harjavalta, southwest (SW) Finland, was simulated. Treatment I (control) represented approxi mately the same concentration of nutrients and treat ment II about the same concentration of nutrients and heavy metals as rainwater at a distance of 0.5 km from the smelter reported by Helmisaari et ai. (1994) and Helmisaari (personal communication). Nitrate and ammonium nitrogen were applied in the ratio 30:70. In treatments lII—VI, the concentrations of all the components increased exponentially, the relative concentrations thus being constant in the different treatments. The solutions contained macronutrients Mg, Ca, P, K, N and heavy metals Cu, Ni, Pb, Zn, Fe, Cd, Cr, Mn including also plant essential trace elements (Table 1). Contamination of the substrate and possible element uptake via the roots were avoided by protecting the roots of the seedlings with a plastic cover on each pot. The contamination of pot surface via the stem flow of E. nigrum was avoided by fastening the stem to plastic sheet with sticky tack. Discoloured leaves were removed to ensure that all the leaves of the plants were alive at the beginning of the experiment. The aerial parts of the seedlings were enclosed in a plastic container, the spraying treatment being applied via openings in the Table 1 The experimental setup Element Main substance used Element conc. (mg l -1 ) in spraying solution for different treatments I II III rv V VI Cu CuS0 4 0 0.4 1.2 3.7 11.4 35.0 Ni NiCl 2 0 0.05 0.2 0.5 1.4 4.4 Pb PbCl2 0 0.02 0.05 0.1 0.4 1.3 Zn ZnS04 x7H2 0 0 0.05 0.1 0.4 1.3 3.9 Fe FeS0 4 x7H 2 0 0 0.2 0.5 1.4 4.3 13.1 Cd 3CdS0 4 x8H 2 0 0 0.002 0.005 0.01 0.04 0.1 Cr CrCl 2 0 0.02 0.05 0.1 0.4 1.3 Mn MnS0 4 xH 2 0 0.005 0.005 0.02 0.05 0.1 0.4 Mg MgCl2 x6H 20 0.07 0.07 0.2 0.6 2.0 6.1 Ca CaCl 2 x2H 2 0 0.2 0.2 0.6 1.9 5.7 17.5 P kh 2po4 0.02 0.02 0.06 0.2 0.6 1.8 K KC1 0.1 0.1 0.3 0.9 2.9 8.8 N (NO3.NH4, 30:70) NRjNOJ. NRjCl 0.9 0.9 2.6 7.9 24.2 74.4 S. Monni et ai. / Environmental Pollution O (2000) 1-10 3 container. The seedlings were sprayed from two opposite directions with 15 ml of solution from each direction (altogether 15 times). There were nine plants per treatment. 2.2. Chemical analysis of the soil and plant material In order to determine possible contamination of the substrate, the surface peat was removed from the pots at the end of the experiment, dried at +6O°C and passed through a 0.4-mm sieve. The total concentrations of Ca, Cd, Cu, Fe, K, Mg, Mn, Ni, P, Pb and Zn in the peat were determined by dry digestion (+55O°C), followed by extraction of the ash with 2-3 ml of 6 M HCI (pro analysi) at approximately +BO°C. The dry residue was dissolved in 10 ml of 1 M HCI for 20 min and filtered (filter paper; Schleicher & Schuell 589 3 ) with 0.1 M HCI. The solutions were analysed by induction coupled plasma atomic emission spectrometry (ICP-AES). After 7 weeks of treatment, the plants were harvested, divided by year growth, rinsed with distilled water for 1 min to minimize the effect of surface contamination, and then oven-dried at + 60° C. The plants were divided into the following fractions: (1) leaves of current-year growth; (2) leaves of previous-year growth; (3) older leaves; (4) stems of current-year growth; (5) stems of previous-year growth; (6) older stems; (7) bark of current-year stems; (8) bark of previous-year stems; and (9) bark of older stems. In control treatments, the bark of current- and previous-year stems were not separated. The different parts were homogenized, and the element concentrations determined by dry digestion (+55O°C) as described above. Total concentrations Of Ca, Cd, Cu, Fe, K, Mg, Mn, Ni, P, Pb and Zn were analysed by ICP-AES. The C and N concentrations were determined from the dry material using a CHN Leco analyser (Nelson and Sommers, 1982). 2.3. Analysis of chlorophyll For the chlorophyll analyses samples of the current year growth of E. nigrum were collected in liquid nitrogen in the greenhouse and the samples stored at —BO°C. The shoots were freeze-dried (Hetosicc freeze dryer, type CD 52), leaves and stems separated and leaves ground with a mortar and pestle. Aliquots of 10 mg were extracted in 3 ml of 80% acetone over night at + 4°C, centrifuged for 3 min, and the absorbances measured at 647 and 664 nm on a spectrophotometer (Shimadzu UV-1201, UV-VIS). Two replicates, each based on a composite sample of up to four plants, were analysed for all the treatments. Chlorophyll a (nmol l -1 ) from leaves was calculated according to equation 13.19x/4 and chlorophyll b (|imol 1 _1 ) according to the equation HAOxA^—s.26xA 664 (Graan and Ort, 1984). 2.4. CO2 exchange rate Before the plants were harvested, the C02 exchange was measured using a battery-operated Li-Cor LI-6200 portable photosynthesis system (Li-Cor Inc., Lincoln, NE, USA). Measurements were carried out with a quarter-litre chamber. A quantum sensor was situated outside and temperature and relative humidity sensors inside the chamber. The portable light source consisted of one halogen lamp (type; electro-valo, EV-KHV-400, Na-lamp, agro, 400 W). The ambient C02 concentra tion in the greenhouse was between 360 and 399 ppm. Readings were collected at 2-s intervals for 42. Six dif ferent irradiance levels (0, 50, 110, 330, 600 and 820 |imol m~ 2 s~ l ) of photosynthetically active radiation (PAR) were used. Before the measurements, the plants were acclimatized at each light level for 5 min. All six irradiance levels were applied to the same plants. Five plants per treatment were measured. In each measure ment, two to three branches of the current-year growth were placed in the chamber, the number of leaves inside the chamber being about 200. Dark respiration was measured by covering the chamber with a black plastic sheet. After taking measurements, all leaves were removed from the measured stems, the leaf area of a single leaf estimated using the formula of an ellipse and the total leaf area calculated. 2.5. Stem water potential The total water potential of the xylem sap of E. nigrum was measured with a Scholander pressure bomb (Scholander et al., 1965) at the end of the experiment. The stems of the previous-year growth were cut with a razor blade and the stem water potential was measured in normal room light. Five plants in each treatment and two pieces of stem from each plant were measured. 2.6. Analysis of ABA For ABA analyses the current-year shoots of E. nigrum were collected at the end of the experiment (up to four plants in the composite sample per treatment). The samples were frozen in liquid nitrogen and stored at—Bo°C. The shoots were freeze-dried (Hetosicc freeze dryer, type CD 52), leaves and stems separated and ground with a mortar and pestle. One-hundred milli grams of homogenized plant material was suspended in 5 ml of 0.05 M phosphate buffer (pH 8.0), and 50 ng internal standard ([ 2 H 4]ABA) was added. After extrac tion in darkness under shaking at 4°C for a minimum of 2 h, the samples were centrifuged at 5000 rpm for 15 min. The supernatant was removed, and the pellet was resuspended in 2 ml of 0.05 mM phosphate buffer (pH 8.0) and centrifuged at 5000 rpm for 15 min. The supernatants were combined, the pH was adjusted to 2.7 4 S. Monni et al. /Environmental Pollution O (2000) 1-10 with 0.1 M HCI, and partitionated three times against equal volumes of ethyl acetate. The pooled ethyl acetate phase was reduced to dryness in vacuo in a rotary eva porator at 40° C. The residue was dissolved in 100 nl of 80% methanol, injected into a Ci B high-performance liquid chromota graphy (HPLC)-column, and eluated with a gradient of 30-50% methanol in 1% aqueous acetic acid over 20 min. The technical specifications of the HPLC system have been described earlier (Monni et al., 2000b). The fraction containing ABA was collected, methylated using ethereal diazomethane, and analysed by GC-MS as described by Monni et al. (2000b). The MS source Fig. 1. Copper concentrations (mg kg -1 ) in the (a) leaves, (b) stems and (c) bark of different year classes of Empetrum nigrum (values are based on one combined sample of six to nine plants per treatment). was operated in electron impact mode at 70 eV, and ions of m/z 162, 190, 193 and 194 were monitored. 2.7. Statistical analysis Non-parametric Kruskal-Wallis analysis of variance (ANOVA) was used in analysing the effects of spraying treatments on the stem water potential of E. nigrum. Pairwise comparisons between the treatments were per formed by the Kruskal-Wallis, comparison of mean ranks test. The Spearman rank correlations were calcu lated between Cu and Fe concentrations in different plant parts and stem water potential, chlorophyll and ABA contents of E. nigrum leaves. To see the element uptake via the roots from the peat and possible con tamination of peat by spraying solutions, the Spearman rank correlations between the element concentrations in peat and different parts of E. nigrum were calculated (Sokal and Rohlf, 1995; Statistix, 1996). The effect of treatment on C02 exchange rate of E. nigrum was evaluated by comparing the dark respiration at the irradiance level of 0 (imol m~ 2 s _l and the max imum photosynthesis at the irradiance levels of 600 or 820 (imol m~ 2 s -1 in different treatments. One-way ANOVA was performed and the differences between the treatment means were compared by (-test (SAS Institute Inc., 1994). 3. Results 3.1. Chemical analysis of the surface peat and plant parts Increased Ca, K, Mg and P concentrations were found in the peat in treatment 111. The Cu and Zn con centrations were also higher in treatments 111 and VI than in the other treatments, indicating that peat con tamination could not be completely prevented (Table 2). However, there was no systematic correlation of those elements in peat and plant parts indicating that the increased concentrations in peat did not affect clearly on the element concentrations in the plant parts. Table 2 Element concentrations (mg kg -1 ) in the surface peat (six to nine plant substrates in one bulk sample/treatment)" a below the analytical detection limit. Treatment Ca Cu Cd Fe K Mg Mn Ni P Pb Zn I 6779 81 _ 2396 870 2790 48 2.4 410 3.9 51 II 8 362 84 - 2056 882 3095 50 - 459 5.9 45 III 11052 116 - 2305 1368 4298 58 2.6 685 4.7 63 IV 8 585 101 - 2394 1163 3477 61 2.5 552 4.8 56 V 6734 94 - 2401 996 2802 55 2.6 465 4.4 60 VI 6 538 120 - 1774 1038 2620 44 - 519 6.0 73 S. Monni et ai. / Environmental Pollution O (2000) 1-10 5 Table 3 Element concentrations and C/N ratio in the (1) current-, (2) previous-year and (3) older leaves of Empetrum nigrum (six to nine plants in one bulk sample/treatment)a a below the analytical detection limit. Cu concentrations in the leaves, stems and bark of all the years increased with increasing Cu concentrations in the spraying solutions (Fig. 1). The highest Cu con centrations occurred in the bark (2147 mg kg -1) (Fig. lc). The maximum concentration of Cu in the leaves was 260 mg kg -1 (Fig. la) and in the stems 80 mg kg -1 (Fig. lb). Cu accumulation increased in older leaves and bark (Fig. la, c). In the stems, Cu concentrations were higher in older and current-year stems than in the pre vious-year stems (Fig. lb). The concentrations of other heavy metals in the plant parts also increased due to increasing heavy metal con centrations in the spraying solution. Iron concentrations increased in all plant parts except in the previous-year and older stems (Tables 3-5), and the highest Fe con centrations occurred in the older bark (659 mg kg -1 ) Table 4 Element concentrations (mg kg -1 ) in the (1) current-, (2) previous-year and (3) older stems of Empetrum nigrum (six to nine plants in one bulk sample/treatment)a • below the analytical detection limit. Year growth Treatment Ca Cd Fe K Mg (mg kg -') Mn Ni P Pb Zn C N (% dry matter) C/N 1 I 6084 _ 43 6080 2353 321 _ 1661 _ 28 54.7 1.28 43 1 II 6003 - 44 7179 2483 330 - 1921 28 54.6 1.25 44 1 III 5081 - 42 7929 2227 281 - 1895 f _ " :-- 28 55.0 1.30 42 1 IV 5683 - 62 7310 2218 297 - 1570 27 55.2 1.17 47 1 V 5959 - 95 6572 2226 274 3.5 1484 25 54.9 1.08 51 1 VI 5602 - 188 7194 2284 289 5.0 1596 9.5 29 54.8 1.13 49 2 I 9324 - 56 4186 3092 639 - 1561 - ... 43 53.8 1.25 43 2 II 8645 - 91 3999 2871 648 3.8 1423 41 53.9 1.19 45 . 2 III 7726 - 68 4644 2813 575 - 1403 - 42 54.0 1.28 42 2 IV 8014 - 83 4284 2756 622 3.8 1306 - 40 54.5 1.14 48 2 V 9433 - 133 3316 3016 617 . . - • 1312 .. - 37 53.9 1.03 52 2 VI 8372 - 251 4537 2938 585 7.0 1335 11.9 40 53.8 1.12 48 I 16401 - 90 2665 4530 1494 - 1160 - 66 52.6 0.98 54 II 14824 - 79 2941 4198 1539 : - 1312 - 61 52.5 1.03 51 III 15262 - 80 3094 4455 1424 5.2 1126 - 64 52.1 1.01 52 IV 13737 0.3 99 2935 4068 1321 5.4 1149 4.3 55 53.2 0.97 55 V 12461 0.3 136 2998 3743 987 7.9 1121 6.3 48 52.7 0.84 63 VI 14702 0.6 223 3267 4430 1527 16.8 977 12.3 75 52.7 0.86 61 Year growth Treatment Ca Cd Fe K Mg Ma Ni F Pb Zn 1 I 3439 - 30 3075 1223 373 - 8 2 4-25 1 II 2658 - 59 2968 1085 321 5.6 8 1 2-23 1 III 3160 - 36 3940 1283 355 - 8 9 5 - 26 1 IV 3496 - 135 2996 1251 396 16.3 7 9 6-29 1 V 3048 - 74 2859 1117 331 4.9 7 9 2 - 22 I VI 2533 - 129 2947 969 306 5.4 7 8 4 6.1 24 2 I 1469 _ 202 1838 558 296 28.4 7 4 1 - 14 2 II 1244 - 42 1878 470 258 - 6 8 7 - 12 2 III 1496 - 29 2369 560 265 - 6 8 3 - 16 2 IV 1721 - 79 1951 588 357 10.7 7 2 0-16 2 v 1412 37 1623 551 290 6 5 8-14 2 VI 1439 - 43 1942 552 256 3.4 6 8 3 - 15 3 I 1530 _ 70 1564 562 403 7.5 6 16 - 15 3 n 1540 - 36 1693 588 385 - 6 3 m 1414 - 36 1924 508 345 3.0 5 81 - 20 3 IV 1491 - 39 1858 579 405 3.4 6 47 - 19 3 V 1514 0.4 62 1691 638 395 6.2 6 08 - 18 3 VI 1504 0.4 49 1711 528 385 6.3 6 14 - 21 6 S. Monni et ai. / Environmental Pollution O (2000) 1-10 (Table 5). Pb and Ni concentrations increased in the previous-year and older leaves and bark, especially. Ni concentrations, however, remained low apart from the older bark where the maximum concentration was 164 mg kg -1 (Table 5). Cd was only detected in older tissues (Tables 3-5), primarily in the bark, where Cd levels increased with increasing metal applications (Table 5). Stem Zn levels were not affected by treatments or tissue age (Table 4). Foliar Zn and Mn concentrations increased with leaf age, but this was not related to the applied levels (Table 3). In the bark, Zn levels were the highest in the oldest bark and increased with increasing applications (Table 5). Macro-nutrient concentrations did not increase in any plant parts in response to increasing nutrient applica tions. However, concentrations of K and Mg in older bark and concentrations of Mg in current-year stems, slightly decreased with increasing nutrients in spraying solutions (Tables 4 and 5). The P concentrations decreased in previous-year and older leaves, and N concentrations in the leaves, with increasing nutrient applications. Foliar N concentrations decreased and the C/N ratio increased with increasing age. In addition, the C/N ratio increased slightly with increasing treatment level (Table 3). The Ca and Mg concentrations increased and the K and P concentrations decreased with increasing tissue age in the leaves (Table 3), while Ca, K, Mg and P decreased with increasing tissue age in the stems of E. nigrum (Table 4). In the bark, Ca concentration increased and K and Mg concentrations decreased with tissue age, while the age of the bark had no clear effect on P concentrations (Table 5). 3.2. Chlorophyll concentration The sum of chlorophyll a and b did not change with increasing heavy metal and nutrient concentrations in the spraying solution (Fig. 2a). The highest chlorophyll concentrations were found in the leaves from treatments I, IV and VI, and the lowest in the plants from treat ment V. The sum of chlorophyll a + b correlated posi tively with the Cu concentrations in the current-year bark and Fe concentrations in the previous-year stems (/■, = 0.94-1.00, P < 0.05). The chlorophyll alb ratio increased slightly with increasing heavy metal and nutrient levels (Fig. 2b). The Cu concentrations in the previous-year stems, current-year and older bark and Fe concentrations in older leaves and current- and pre vious-year bark correlated positively with the chlor ophyll a/b ratio (r,= 0.83-1.00, P < 0.05). 3.3. C02 exchange The highest C0 2 exchange rate of E. nigrum was measured at the irradiance levels of 600 and 820 jimol m -2 s -1 (Fig. 3a). The increasing element concentra tions in the spraying solution decreased dark respiration (Fig. 3b) and maximum photosynthesis of E. nigrum current-year shoots (Fig. 3c). The dark respiration was the lowest in treatments of V and VI and these differed statistically from the treatments I, II and 111 (P<0.01; F= 5.27; Fig. 3b). Maximum photosynthesis was the lowest in the highest treatments (V and VI) but these did not differ statistically from the other treatments (P = 0.0798, F= 2.27; Fig. 3c). 3.4. Water potential The mean water potential of E. nigrum varied between—ll and —l3 bars. The increasing treatment levels had no consistent effects on the water potential (Fig. 4a). The water potential correlated negatively with the Cu or Fe concentrations in the current- and pre vious-year leaves and Cu concentrations in older stems and previous-year bark (r, = 0.83-0.90, P< 0.05). 3.5. ABA contents The ABA contents varied between 13 and 66 ng g -1 dry wt. The spraying treatment appeared to affect the ABA contents, the highest contents being in treatments IV and VI. The lowest values were measured in treat Fig. 2. (a) Chlorophyll a+b concentrations (µmol 1-1 ) and (b) chlor ophyll a/b ratio in the current-year leaves of Empetrum nigrum. Results are based on one combined sample of up to four plants per treatment. Each sample was measured twice. S. Monni et ai. I Environmental Pollution O (2000) 1-10 7 Table 5 Element concentrations (mg kg -1 ) in the (1) current-, (2) previous-year and (3) older bark of Empetrum nigrum (six to nine plants in one bulk sample/treatment).a a below the analytical detection limit. ments II and 111 (Fig. 4b). The Cu and Fe concentrations in older leaves, Cu concentrations in the current-year stems and bark and Fe concentrations in older bark correlated positively with ABA content (r, = 0.83-0.94, P< 0.05). 4. Discussion The accumulation of Cu, Fe, Pb, Cd, Ni and Zn in leaves or bark increased along with increasing con centrations of heavy metals in the spraying solution. In contrast, there was no change in the concentrations of macronutrients and Mn. This difference may be partly related to differences in the surface binding of the dif ferent elements. Atteia and Dambrine (1993) reported that the Ni, Zn and Pb concentrations were unaffected and that Fe was retained by the needles when rainfall passed down through the tree crowns. Fe was also retained on the surface of spruce twig axes, even though the twigs were washed with toluene and tetrahydrofuran (Wyttenbach et al., 1988). In contrast, Little (1973) suggested that a high proportion of Zn, Pb and Cd can be removed by deionised water from broadleaved spe cies. However, surface contamination may significantly contribute to the levels of heavy metals in the above ground parts of plants exposed to aerial pollution (Alfani et al, 1996). In contrast, macronutrients are easily washed from leaf and bark surfaces, which is supported by earlier results with leaves (Cercasov et al., 1987) and stems (Wyttenbach et al., 1988). Wyttenbach et al. (1988) found that washing resulted in the removal of particulate material containing Ca, K, Mg, Mn, Na and P from the surface of bark. The results of this study showed that the leaves and bark of E. nigrum have a high capacity to bind heavy metals. Furthermore, this capacity seems to increase with increasing age of the tissue. If accumulation takes place primarily at the sur face it may have only limited physiological effects on E. nigrum, but the abscission of older leaves and bark tis sue will result in significant accumulation of heavy metals in the soil under the plants. Increasing heavy metal concentrations with age in the leaves and stems of E. nigrum growing in the field and greenhouse have been found, and suggests that E. nigrum accumulates metals in the older tissues (Helmi saari et ai, 1995; Uhlig et ai., 1996; Monni et ai., 2000 a). The accumulation pattern was similar whether the metals were applied to the roots (Monni et ai., 2000 a) or to the aboveground parts in the greenhouse. Therefore, these studies suggest that both accumulation and surface contamination contribute to the high metal concentrations in the older parts of E. nigrum in the field. On the contrary, there was no age-dependant accumulation in the stems; Cu, Zn and, to some extent, the Fe concentrations were higher in the current-year stems than previous-year stems. The bark is very diffi cult to separate from the other stem tissue in current year stems, whereas removal is easier in previous-year and older stems. For this reason, it is very likely that the current-year stem samples contained some bark tissue, thus increasing the concentrations of heavy metals in these samples. The metal concentrations in the different above ground parts of E. nigrum decreased generally in the following order: bark > leaves > stems. In a study where Year growth Treatment Ca Cd Fe K Mg Mn Ni P Pb Zn 1 II 6836 _ 76 2445 1487 346 _ ; 398 _ 46 1 III 6720 - 102 2915 1666 435 - 512 - 42 1 IV 7526 - 100 2494 1635 415 - 422 - 49 1 V 7578 - 120 2103 1526 361 ■ 348 - 43 1 VI 7585 - 249 2349 1539 437 411 - 51 2 II 7446 _ 127 1335 1645 483 366 _ 102 2 III 6956 - 128 1733 1430 456 /•;. - 357 - 65 2 IV 7338 - 87 1272 1442 495 : - 308 - 62 2 V 7927 - 157 1001 1480 477 - 302 - 61 2 VI 7175 - 243 1259 1435 470 20.3 336 19.9 70 3 I 9626 0.4 181 550 1133 353 5.3 333 6.7 178 3 II 8760 1.2 171 410 1006 317 8.4 289 - 195 3 III 9751 0.7 180 587 1122 341 16.8 365 - 213 3 IV 8664 3.6 188 567 1049 368 30.8 320 7.5 218 3 V 9136 6.1 340 449 992 350 78.8 314 20.5 235 3 VI 10193 5.5 659 380 904 341 163.9 347 51.5 333 8 S. Monni et al. / Environmental Pollution O (2000) 1-10 Fig. 3. (a) The measured values of C02 exchange rate (CO 2 µmol m-2 s -1 ) of Empetrum nigrum current-year shoots in all treatments (n =4- 5/treatment) in varying light intensities of photosynthetically active radiation (PAR; µmol m- 2 s -1 ), (b) the dark respiration (CO2 µmol m-2 s -1 ) at irradiance level of 0 µmol m- 2 s- 1 (n = 5/treatment) and (c) the maximum photosynthesis (CO2 m-2 s -1 ) at irradiance levels of 600 or 820 µmol m-2 s-1 (n = 5/treatment) of Empetrum nigrum in different treatments (P = 0.0798). Bar indicates the standard deviation and different letters statistical differences between the treatment means according to the t-test (P < 0.05). the bark was not removed, metal accumulation was reported to be higher in the stems than in the leaves in field-grown plants (Uhlig et al., 1996). The pattern was the same in a greenhouse experiment, where the metals were applied to the roots and surface contamination was excluded (Monni et al., 2000 a). Helmisaari and Siltala (1989) found that the inner bark (living bark and phloem) of pine acted as a sink for elements, whereas smaller amounts accumulated in the wood and outer Fig. 4. (a) Water potential (—bar) of Empetrum nigrum {n = 9-10). Bar indicates the standard deviation. Different letters indicate statistical differences between the treatment means according to the Kruskall- Wallis, comparison of mean ranks test (P<0.05). (b) Abscisic acid (ABA) content ng g -1 dry weight in current-year leaves of Empetrum nigrum. Results are based on one combined sample up to four plants per treatment. bark (Helmisaari and Siltala, 1989). Because the metals in this study were applied to the aerial parts of E. nigrum, and the stem samples were only divided to bark (including mainly epidermis and cork) and other stem tissue, more specific place in heavy metal accumulation in stem tissue cannot be evaluated. Several studies have shown a negative influence of heavy metals on the photosynthetic pigments (Angelov et al., 1993; Bishnoi et al., 1993). Also more negative water potential by heavy metal treatments has been found (Bishnoi et al., 1993). However, in this experiment, the aerial application of heavy metals had no clear effect on the water potential and chlorophyll contents of£. nigrum. Chlorophyll a/b ratio, which is used as a stress indicator, increased slightly with increasing metal treatments, which was also seen in E. nigrum leaves near the Cu-Ni smelter in the field (Monni et al, 2000b). The chlorophyll a/b ratio has been reported to increase due to environmental stress (Delfine et al., 1999). The ABA contents increased slightly with increasing metal treatments, which has also been found earlier in experimental studies. In bean leaves the ABA content increased two-fold due to excess Ni (Rauser and Dumbroff, 1981), which is about as high increase as in E. nigrum leaves when exposed to the S. Monni et al. / Environmental Pollution O (2000) 1-10 9 highest heavy metal treatment. The overall ABA contents of E. nigrum leaves were lower than the values measured in the field (Monni et al., 2000b) and in conifer trees (Tan and Blake, 1993; Yang et al., 1993; Bianco and Dalstein, 1999). The response of respiration of E. nigrum to tempera ture has been measured before and found to increase with increasing temperature (Wager, 1941). Otherwise, CO2 exchange of E. nigrum has not been reported, but the observed rates were about the same level as those of Pinus sylvestris in Europe (Luoma, 1997). There was also an indication of decrease in dark respiration and maximum photosynthesis due to increasing treatment levels. The decrease in net photosynthesis due to heavy metals is supported by earlier studies (Lamoreux and Chaney, 1978; Angelov et al., 1993; Bishnoi et al., 1993). In contrast to this study, Lamoreux and Chaney (1978) found an increase in dark respiration by Cd, though the dark respiration was not clearly correlated with Cd content in tissue. The total N and C concentrations in the leaves of E. nigrum were about the same level as those reported for the leaves and shoots of Empetrum hermaphroditum growing in North Scandinavia (Maimer and Nihlgärd, 1980; Michelsen et al., 1996 a, b). The increase in the C/ N ratio with increasing treatment levels was due to the decreasing N concentrations in the leaves, and is contrary to the results of Michelsen et al. (1996 a), who found that NPK fertilizers increased the N concentra tions in shoots. They also found that fertilizers had no effect on the chlorophyll contents of E. hermaphroditum (Michelsen et al., 1996 a). In the study of Michelsen et al. (1996 a), however, the fertilizers were applied to the roots and the plants were not affected by heavy metal pollution. No correlation was found between leaf N concentrations, CO2 exchange and chlorophyll a and b concentrations, although an increased N supply gen erally increases the photosynthetic rate (Hoogesteger and Karlsson, 1992). These results are consistent with the suggestion that the applied macronutrients were not taken up by the aboveground parts of E. nigrum. 5. Conclusions The results showed that Cu, Fe, Pb, Cd, Ni and Zn applied to the aerial parts, were strongly accumulated on the bark and leaves of E. nigrum, while the applied macronutrients and Mn were readily removed from the leaf and bark surfaces by distilled water. Therefore, metal and macronutrient uptake by E. nigrum leaves, as reported earlier for crops (Chamberlain, 1983) and trees (Lin et al., 1995; Koricheva et al., 1997), was not con firmed by this study. No detectable changes in the water potential and chlorophyll concentrations were observed, indicating that the treatments had no critical effects on the ecophysiology of E. nigrum. This is consistent with the suggestion that the applied metals were located pri marily outside the symplast. However, the surface con tamination was found to decrease the dark respiration. Also an indication of a decrease in the maximum pho tosynthesis and an increase in the leaf ABA content of E. nigrum was found. Acknowledgements We would like to thank the staff at the Finnish Forest Research Institute, at the Ruotsinkylä greenhouse and at the University of Tromso for helping with the prac tical work. Thanks go to Maarit Martikainen, Kerttu Nyberg and Pirkko Ronkainen at the laboratory of the Finnish Forest Research Institute for performing the chemical analyses and to Jargen Melmann at the Uni versity of Tromsß for measuring ABA contents. We would also like to thank Jarkko Koskela at the Uni versity of Helsinki for his advice on the use of the Licor photosynthesis system. Jarkko Koskela and Maija Salemaa made critical comments on the manuscript and John Derome revised the English. The work was funded by the Maj and Tor Nessling Foundation, the Alfred Kordelin Foundation and NorFA. References Alfani, A., Maisto, G., lovieno, P., Rutigliano, F.A., Bartoli, G., 1996. Leaf contamination by atmospheric pollutants as assessed by elemental analysis of leaf tissue, leaf surface deposit and soil. Jour nal of Plant Physiology 148, 243-248. Angelov, M., Tsonev, T., Uzunova, A., Gaidardjieva, K., 1993. Cu 2 + effect upon photosynthesis, chloroplast structure, RNA and protein synthesis of pea plants. Photosynthetica 28, 341-350. Antonovics, J., Bradshaw, A.D., Turner, R.G., 1971. Heavy metal tolerance in plants. Advances in Ecological Research 7, 1-85. Atteia, 0., Dambrine, E., 1993. Dynamique d'elements traces dans les precipitations sous le couvert de 2 pessieres peu polluees de Suisse romande. Annales des Sciences Forestieres 50, 445-459. Baker, A.J.M., 1987. Metal tolerance. New Phytologist 106, 93-111. Becerril, J.M., Munoz-Rueda, A., Aparicio-Tejo, P., Conzalez-Murua, C., 1988. The effects of cadmium and lead on photosynthetic electron transport in clover and lucerne. Plant Physiology and Bio chemistry 26, 357-363. Becerril, J.M., Conzalez-Murua, C., Munoz-Rueda, A., De Felipe, M.R., 1989. Changes induced by cadmium and lead in gas exchange and water relations of clover and lucerne. Plant Physiology and Biochemistry 27, 913-918. Bell, J.N.8., Tallis, J.H., 1973. Biological flora of the British Isles. Empetrum nigrum L.. Journal of Ecology 61, 289-305. Bianco, J., Dalstein, L., 1999. Abscisic acid in needles of Pinus cembra in relation to ozone exposure. Tree Physiology 19, 787-791. Bishnoi, N.R., Sheoran, 1.5., Singh, R., 1993. Influence of cadmium and nickel on photosynthesis and water relations in wheat leaves of different insertion level. Photosynthetica 28, 473-479. Carlquist, S., 1989. Wood and bark anatomy of Empetraceae; com ments on paedomorphosis in woods of certain small shrubs. Aliso 12, 497-515. 10 S. Monni et ai. / Environmental Pollution 0 (2000) 1-10 Cercasov, V., Rentschler, 1., Schreiber, H., 1987. Deposition of heavy metals and their distribution on needles and bark. In: Mathy, P. (Ed.), Air Pollution and Ecosystems. Proceedings of an Interna tional Symposium, 18-22 May, Grenoble, pp. 766-773. Chamberlain, A.C., 1983. Fallout of lead and uptake by crops. Atmospheric Environment 17, 693-706. Chertov, 0.G., Nadporozshskaya, M.A., Lapshina, 1.N., Grigorieva, 0.A., 1993. Soil degradation in surroundings of "Pechenganikel" smelter complex. In: Kozlov, M.V., Haukioja, E., Yarmishko, V.T., (Eds.), Aerial Pollution in Kola Peninsula. Proceedings of the International Workshop, 14-16 April, St Petersburg, pp. 159-162. Delfine, S., Alvino, A., Villani, M.C., Loreto, F., 1999. Restrictions to carbon dioxide conductance and photosynthesis in spinach leaves recovering from salt stress. Plant Physiology 119, 1101-1106. Good, R.D'O., 1927. The genus Empetrum L. Journal of the Linnean Society of London 47, 489-523. Graan, T., Ort, D.R., 1984. Quantitation of the rapid electron donors to P7OO, the functional plastoquinone pool, and the ratio of the photosystems in spinach chloroplasts. Journal of Biological Chem istry 259, 14003-14010. Helmisaari, H.S., Siltala, T., 1989. Variation in nutrient concentra tions of Pinus sylvestris stems. Scandinavian Journal of Forest Research 4, 443-451. Helmisaari, H.S., Derome, J., Fritze, H., Nieminen, T., Palmgren, K., Salemaa, M., Vanha-Majamaa, 1., 1995. Copper in Scots pine for ests around a heavy-metal smelter in South-Western Finland. Water, Air and Soil Pollution 85, 1727-1732. Helmisaari, H.-S., Derome, J., Levula, T., Nieminen, T., Palmgren, K., Salemaa, M., Vanha-Majamaa, 1., Reponen, P., Rutter, 0., 1994. Metsikön ravinnekierto ilman epäpuhtauksien aiheuttaman kuormituksen osoittajana. In: Mälkönen, E., Sivula, H. (Eds.), Suomen metsien kunto. Metsäntutkimuslaitoksen tiedonantoja 527, pp. 156-174. Hoogesteger, J., Karlsson, P.S., 1992. Effects of defoliation on radial stem growth and photosynthesis in the mountain birch (Betula pub escens ssp. tortuosa). Functional Ecology 6, 317-323. Koricheva, J., Roy, S., Vranjic, J.A., Haukioja, E., Hughes, P.R., Hänninen, 0., 1997. Antioxidant responses to simulated acid rain and heavy metal deposition in birch seedlings. Environmental Pol lution 95, 249-258. Laaksovirta, K., Silvola, J., 1975. Effect of air pollution by copper, sulphuric acid and fertilizer factories on plants at Harjavalta, W. Finland. Annales Botanici Fennici 12, 81-88. Lamoreux, R.J., Chaney, W.R., 1978. The effect of cadmium on net photosythesis, transpiration, and dark respiration of excised silver maple leaves. Physiologia Plantarum 43, 231-236. Lin, Z.Q., Barthakur, N.N., Schuepp, P.H., Kennedy, G.G., 1995. Uptake and translocation of 54 Mn and 63 Zn applied on foliage and bark surfaces of balsam fir {Abies balsamea (L.) Mill.) seedlings. Environmental and Experimental Botany 35, 475-483. Little, P., 1973. A study of heavy metal contamination of leaf surfaces. Environmental Pollution 5, 59-172. Luoma, S., 1997. Geographical pattern in photosynthetic light response of Pinus sylvestris in Europe. Functional Ecology 11, 273-281. Mälkönen, E., Derome, J., Fritze, H., Helmisaari, H.S., Kukkola, M., Kytö, M., Saarsalmi, A., Salemaa, M., 1999. Compensatory fertili zation of Scots pine stands polluted by heavy metals. Nutrient Cycling in Agroecosystems 55, 239-268. Maimer, N., Nihlgärd, 8., 1980. Supply and transport of mineral nutrients in a sub-arctic mire. In: Sonesson, M. (Ed.), Ecology of a Subarctic Mire, Ecological Bulletins 30, 63-95. Michelsen, A., Jonasson, S., Sleep, D., Havström, M., Callaghan, T.V., 1996 a. Shoot biomass, 8 13 C, nitrogen and chlorophyll responses of two arctic dwarf shrubs to in situ shading, nutrient application and warming simulating climatic change. Oecologia 105, 1-12. Michelsen, A., Schmidt, 1.K., Jonasson, S., Quarmby, C., Sleep, D., 1996b. Leaf ,5 N abundance of subarctic plants provides field evi dence that ericoid, ectomycorrhizal and non- and arbuscular mycorrhizal species access different sources of soil nitrogen. Oeco logia 105, 53-63. Monni, S., Salemaa, M., Millar, N., 2000 a. The tolerance of Empetrum nigrum to copper and nickel. Environmental Pollution 109, 221-229. Monni, S., Uhlig, C., Hansen, E., Magel, E., 2000b. Ecophysiological responses of Empetrum nigrum to heavy-metal pollution. Environ mental Pollution, , (in press). Nelson, D.W., Sommers, L.E., 1982. Total carbon, organic carbon and organic matter. In: Page et al. (Eds.), Methods of Soil Analysis, part 11, 2nd Edition. Soil Science Society of America, Madison, Wisconsin, pp. 574-577. Proctor, J., Woodell, S.R.J., 1971. The plant ecology of serpentine (I. Serpentine vegetation of England and Scotland). Journal of Ecology 59, 375-395. Rauser, W., Dumbroff, E., 1981. Effects of excess cobalt, nickel and zinc on the water relations of Phaseolus vulgaris. Environmental and Experimental Botany 21, 249-255. Sarvas, R., 1937. Havaintoja kasvillisuuden kehityksestä Pohjois-Suo men kuloaloilla. Silva Fennica 44, 1-64. SAS Institute Inc, 1994. SAS/STAT User's Guide. Version 6, 4th Edition, SAS Institute Inc, Cary, NC. Scholander, P.F., Hammel, H.T., Bradstreet, E.D., Hemmingsen, E.A., 1965. Sap pressure in vascular plants. Science 148, 339 346. Sokal, R.R., Rohlf, F.J., 1995. Biometry, W. H. Freeman, New York. Statistix, 1996. Statistix for Windows. User's Manual, Borland Inter national Inc, Tallahassee, FL. Tan, W., Blake, T.J., 1993. Drought tolerance, abscisic acid and elec trolyte leakage in fast- and slow-growing black spruce (Picea mari ana) progenies. Physiologia Plantarum 89, 817-823. Uhlig, C., Salemaa, M., Vanha-Majamaa, 1., 1996. Element distribu tion in Empetrum nigrum microsites at heavy metal contaminated sites in Harjavalta, Western Finland. In: Kopponen, P., Kär enlampi, S., Rekilä, R., Kcrenlampi, L. (Eds.), Bio- and Eco technological Methods in Restoration. Abstracts of an International Advanced Course and Minisymposium, 16-18 December, Kuopio, pp. 21. Wager, H.G., 1941. On the respiration and carbon assimilation rates of some arctic plants as related to temperature. New Phytologist 40, 1-19. Weigel, H.J., 1985. The effect of Cd 2+ on photosynthetic reactions of mesophyll protoplasts. Physiologia Plantarum 63, 192-200, Wollenweber, E., Dörr, M., Stelzer, R., Arriaga-Giner, F.J., 1992. Lipophilic phenolics from the leaves of Empetrum nigrum che mical structures and exudate localization. Botanica Acta 105, 300- 305. Wyttenbach, A., Tobler, L., Bajo, S., 1988. Major and trace elements in the twig axes of Norway spruce and the relationship between twig axis and needles. Trees, 204-212. Yang, C., Wessler, A., Wild, A., 1993. Studies on the diurnal courses of the contents of abscisic acid, 1-aminocyclopropane carboxylic acid and its malonyl conjugate in needles of damaged and unda maged spruce trees. Journal of Plant Physiology 141, 624-626. Reprinted from Environmental Pollution, 112, Monni, S., Uhlig, C., Hansen, E., Magel, E. Ecophysiological responses of Empetrum nigrum to heavy metal pollution, in press. © 2000 with permission from Elsevier Science. IV DTD = 4.1.0 EN PO 200 lp Disk used 0269-7491/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved PII: 50269-7491(00)00 1 25 -1 Environmental Pollution 112 (2000) 1-9 Ecophysiological responses of Empetrum nigrum to heavy metal pollution S. Monni a 'b> *, C. Uhlig c , E. Hansen d , E. Magel e * Vantaa Research Centre. Finnish Forest Research Institute, PO Box 18. FIN-01301 Vantaa. Finland b Department of Ecology and Systematics. University of Helsinki, FIN-00014 Helsinki, Finland c Holt Research Centre, The Norwegian Crop Research Institute, N-9292 Tromso, Norway d Department of Plant Physiology and Microbiology, University of Tromso. N-9037 Tromso, Norway c Physiological Ecology of Plants, University of Tubingen. Auf der Morgenstelle I. D-72076 Tubingen, Germany Received 13 December 1999; accepted 18 March 2000 "Capsule": Smeller emissions had a negative effect on the ecophysiology o/Empetrum nigrum (crowberry), a species growing on heavy metal contaminated sites. Abstract Chlorophyll, organic (citric and malic acids) and abscisic acid (ABA) contents and stem water potential were measured to indi cate possible physiological effects of heavy metal deposition on Empetrum nigrum L. (crowberry). The leaves and stems of E. nigrum were collected at distances of 0.5 and 8 km from the Cu-Ni smelter at Harjavalta, south-west Finland. All the investigated parameters were clearly affected by heavy metal emissions. Chlorophyll contents in the leaves and organic acid contents in the leaves and stems were lower close to the emission source. Generally found increase in organic acid contents with increasing Ni concentrations was not found, which might be due to the lower production of organic acids measured by decreased photosynthesis near the smelter. In contrast, ABA contents in stems and leaves in general, were higher in plants growing 0.5 km from the pollution source. Close to the smelter the stem water potential of E. nigrum was less negative during the day but more negative during the night. These results suggest that smelter emissions have a negative effect on the ecophysiology of E. nigrum even though it is considered to be a tolerant species to heavy metals. © Elsevier Science Ltd. All rights reserved. Keywords: Empetrum nigrum L.; Chlorophyll; Organic acids; Abscisic acid, ABA; Stem water potential; Cu-Ni smelter X. Introduction Heavy metal emissions are reported to have serious impacts on plants growing in the surroundings of Cu- Ni smelters (Amiro and Courtin, 1981; Helmisaari et ai., 1995; Shevtsova, 1998). Many dwarf shrub species are described to grow on severely heavy metal contaminated sites in the vicinity of Cu-Ni smelters in the northern hemisphere (Laaksovirta and Silvola, 1975; Bagatto and Shorthouse, 1991; Shorthouse and Bagatto, 1995; Shevtsova, 1998; Mälkönen et ai., 1999). The evergreen dwarf shrub Empetrum nigrum L. (crowberry) has a wide ecological amplitude (Bell and • Corresponding author. Vantaa Research Centre, Finnish Forest Research Institute, PO Box 18, FIN-01301 Vantaa, Finland. Tel.: + 358-9-857-051; fax: +358-9-85705569. E-mail address: satu.monni@metla.fi (S. Monni). Tallis, 1973; Elvebakk and Spjelkavik, 1995), and it survives at a distance of 0.5 km from the Cu-Ni smel ter at Harjavalta, south-west (SW) Finland, where almost all other plant species have disappeared (Laak sovirta and Silvola, 1975; Helmisaari et ai., 1995). At this site, not only the elevated heavy metal concentra tions in the soil (Derome and Lindroos, 1998), but also decreased soil water-holding capacity (Derome and Nieminen, 1998) impairs growing conditions for plants. The metal tolerance mechanism of E. nigrum is not fully known, but one explanation for its survival is the accumulation of high Cu and Ni concentrations in older stems. Thus, E. nigrum is able to restrict the accumulation of Cu and Ni in its younger, growing parts (Helmisaari et ai., 1995; Uhlig et ai., 1996; Monni et ai., 2000). E. nigrum is ericoid mycorrhizal plant and ericoid mycorrhiza of other dwarf shrub species Cal luna vulgaris (L.) Hull, roots have been found to have 2 5. Monni et al. / Environmental Pollution 112 (2000) 1-9 an important role in Cu and Zn tolerance (Bradley et al„ 1981, 1982). The ecophysiological response of E. nigrum to metals has not been previously investigated, but growth reduction in response to increasing emission levels (Shevtsova, 1998) clearly indicates an ecophysiological impact of heavy metals on E. nigrum. In earlier inves tigations, chlorophyll, abscisic acid (ABA) and organic acid contents and water potential of plants have been used to indicate plant response to elevated heavy metal levels (Lee et al., 1978; Rauser and Dumbroff, 1981; Angelov et al., 1993; Bishnoi et al., 1993). The photo synthetic rate and chlorophyll concentrations of plants have been found to be decreased by Cu, Ni, Cd and Pb (Lamoreaux and Chaney, 1978; Becerril et al., 1989; Angelov et al., 1993; Bishnoi et al., 1993; Pandolfini et al., 1996), and an effect of Ni, Zn and Cd on water relations of plants is reported (Rauser and Dumbroff, 1981; Bishnoi et al., 1993). In experimental studies, stomatal conductance and water potential of the Jeaves have decreased and the ABA content, which regulates the water status of the plant, increased when plants were exposed to Ni (Rauser and Dumbroff, 1981; Bishnoi et al., 1993). However, also factors other than heavy metals may influence the ABA relations in the plants; higher levels of ABA in the roots and shoots is a typical response to nitrogen deficiency (Goldbach et al., 1975). ABA is also involved in the synthesis of proteins, prevention of precociuous germination, induction of dormancy and abscission of leaves (Marschner, 1995). In general, plants possess physiological mechanisms that enable them to resist elevated heavy metal con centrations in their substrate (Antonovics et al., 1971; Baker, 1981, 1987; Woolhouse, 1983). Heavy metal tol erance of plants is species- and metal-specific and plants can either detoxify metals by binding them with organic acids, proteins or other ligands (Lee et al., 1978; Rauser and Curvetto, 1980; Godbold et al., 1984), or accumu late the metals in different plant parts or cell organelles (Reilly, 1969; Bringezu et al., 1999). Many plant species which are tolerant to Cu or Ni (Lee et al., 1978; Rauser, 1984) or to both Cu and Ni are found (Hogan and Rauser, 1979). However, the tolerance of these two metals is usually achieved by two different mechanisms. Organic acids play a central role in detoxifying metals in Ni- and Zn-accumulating plants (Ernst, 1975; Mathys, 1977; Lee et al., 1978; Yang et al., 1997), while Cu forms complexes with proteins and amino acids (Rauser, 1984). The aim of this study was to investigate the general ecophysiology of E. nigrum, and to determine the responses of E. nigrum to heavy metal pollution by measuring physiological parameters. Chlorophyll con tents of leaves were used as an indicator for physiologi cal stress, because photosynthesis has been found to decrease due to elevated concentrations of Cu and Ni. Organic acids were measured to find out if there is any connection between organic acid contents and Ni resis tance. To indicate the previously described connection between Cu, Ni and desiccation stress, the stem water potential and abscisic acid contents of E. nigrum leaves and stems were studied. 2. Materials and methods 2.1. The Harjavalta smeller area The shoots of E. nigrum were collected at distances of 0.5 and 8 km to the south-east (SE) of the Cu- Ni smelter at Harjavalta, SW Finland. The Cu smelter was established in 1945 and the Ni smelter in 1960. Sulphur dioxide and heavy metals have been emitted into the environment for the past 40-60 years. The deposition of metals near the smelter was considerably reduced in the 1990s after a taller stack was built and electrostatic filters installed (Rantalahti, 1995). The prevailing wind direction is from the SW, and the emissions are therefore primarily dispersed to the north-east. At both locations (0.5 and 8 km) 10 separate E. nigrum patches were marked situating more than 3 m from each other, and these plants were used for all physiological measurements. The sites were chosen SE from the smelter, because it was the only direction where the two sites represented the same forest site type (Calluna site type) and soil type (orthic podzol). The pH of the organic layer was 3.5 at 0.5 km distance and 3.6 at 8 km distance from the smelter (Derome and Lin droos, 1998). The site at 0.5 km is located in a heavily polluted area where the total Cu and Ni concentrations in the organic layer are over 5800 and 460 mg kg -1 dry wt., respectively, and base cations (e.g. Ca, Mg) have been displaced from the topsoil. The concentrations of other heavy metals (Fe, Zn, Cd, Pb, Cr) in the organic layer are also elevated near the smelter. The site at 8 km is only slightly polluted, the total Cu and Ni concentra tions in the organic layer being 150 and 40 mg kg -1 dry wt., respectively. The concentrations of other heavy metals are also much lower at 8 km than at 0.5 km, but are higher than background values (Derome and Lin droos, 1998). The material collection for physiological measure ments was done on 21 and 26 July, 18-19 and 26 August and 9 October 1997. Near the Cu-Ni smelter at Harjavalta (61° 19' N, 22°9' E), the monthly mean tem perature and precipitation in July were 18.4° C and 86 mm, in August 18.1° C and 38 mm, in September 10.7° C and 111 mm and in October 2.7° C and 66 mm, respec tively, calculated by the model of Ojansuu and Hentto nen (1983). S. Monni et al. / Environmental Pollution 112 (2000) 1-9 3 2.2. Analysis of chlorophyll a and b Several current-year shoots of five to 10 patches of E. nigrum were collected at both distances (0.5 and 8 km) on 21 July, 18-19 and 26 August and 9 October in con tainer of liquid nitrogen (—196 °C), and stored in freezer at -80° C. The current-year shoots were freeze-dried, and separated into leaves and stems and ground with a mortar and pestle. Weighed (10 mg) aliquots were extracted with 3 ml of 80% acetone overnight at + 4°C, centrifuged for 3 min, and the absorbances (A) at 647 and 664 nm recorded on a spectrophotometer (Shi madzu UV-1201, UV-VIS). Two measurements per sample were made. Chlorophyll a (|imol l" 1 ) was calcu lated according to the equation 13.\9xA 66a—2.57xA ml , and chlorophyll b (nmol I - ') according to the equation 22.10 x/( 647-5.26 x/j 6 54 based on MacKinney's coeffi cients (Graan and Ort, 1984), and the results were expressed as chlorophyll content in the tissue (nmol chlorophyll g~' dry weight). 2.3. Analysis of citric and malic acids Several previous-year shoots of five to six E. nigrum patches were collected for organic acid analysis from the two sites. Due to technical reasons the sampling was carried out only on 9 October simultaneously for other sampling. To study the connection between organic acids and heavy metals, the previous-year stems were chosen instead of the most active current-year shoots (as for chlorophyll and ABA), because the metals are accu mulated mainly in the older stems. Immediately after harvesting, the samples were frozen in liquid nitrogen. After freeze-drying, the leaves and stems were separated and ground to a fine powder using a mortar and pestle or a dismembrator (Braun Melsun gen, Germany). Two replicates per sample were prepared. Citric and malic acids were determined enzymatically by a method modified from Boehringer (1989) and Hampp et al. (1984). Twenty milligrams of lyophilized tissue powder and 20 mg of polyvinylpolypyrrolidone (PVPP) were mixed in an eppendorf tube. PVPP was used to minimize the disturbing effect of phenolics, chlorophyll and other compounds. The organic acids were extracted by adding 500 |il 0.1 N HCI, mixing gently for 15 min at room temperature, then incubated at 100° C for 10 min. After cooling, samples were cen trifuged (Hettich microrapid) and aliquots of the super natant were used for the enzymatic determinations. For the citric acid (citrate) analyses 25 nl of extract was incubated with 160 mM glycylglycine buffer (pH 7.8), 0.18 mM ZnCl 2, 0.2 mM NADH and 12 U ml" 1 malate dehydrogenase (all final concentrations) in a total volume of 1 ml. The decrease in optical density after addition 0f0.26U ml - ' citrate lyase was followed at 340 nm in a spectrophotometer (Kontron). The amount of NADH oxidized is stoichiometric with the amount of citrate. In addition, blanks (without sample) and standards containing 0.5 mM citric acid were analysed. For the quantification of malic acid (malate) 25 nl of the extract was incubated with 60 mM glycylglycine buffer (pH 10.0), 9 mM L-glutamate, 4 mM NAD, and 1.8 U ml -1 glutamate-oxaloacetate transaminase (all final concentrations) in a total volume of 1 ml. After addition of 27 U ml -1 malate dehydrogenase, the increase in optical density was recorded at 340 nm and was stoichiometric with the amount of malate present in the sample. Blanks (without sample) and standards containing 1.0 mM of malate were also analysed. 2.4. Stem water potential of E. nigrum and soil moisture Pressure chamber determinations of E. nigrum were carried out to estimate the total water potential of the xylem sap, during the day on 21 July, during the night on 26 July and during the day and night on 18-19 August. The previous-year stems were separated with a razor blade, and the stem water potential measured in sunny or partly cloudy weather using a Scholander's pressure bomb (Scholander et al., 1965; Ritchie and Hinckley, 1975). Night measurements were carried out to determine the relatively constant water potential throughout the plant. Seven to 10 plant patches per site were chosen, and two to three separate stem pieces of each plant measured. Because only one instrument was available, the meas urements could not be performed simultaneously at both sites. Therefore, the water potential measurements of E. nigrum on 21 July were made at noon (12:30- 14:30) at the site at 8 km, and in the afternoon (16:00 17:15) at the site at 0.5 km. The stem water potential measurements of 18 August (at 0.5 km distance) and 19 August (at 8 km distance) were made throughout the day (between 13:00-16:30 in every half an hour) in order to find out the differences between the values obtained at noon and in the afternoon. However, the means of the whole day values were calculated and are shown in the figures because there were no major differences related to the time of the day. The night measurements were done first at 0.5 km distance (12:00-2:20) and then at 8 km distance (3:00-4:40) on 26 July. On 18-19 August the night measurements were also done first at 0.5 km distance (11:20 p.m.-12:40 a.m.) and then at 8 km distance (1:15-2:15). The weather conditions at night were very constant, the air temperature being approxi mately + 12° C and air moisture 20-35% at both sites. The soil moisture measurements were made simulta neously with the stem water potential measurements. The soil moisture was measured using a ThetaProbe soil moisture sensor (type MLI). The ground vegetation was carefully removed and the measurements were made 4 5. Monni et ai. / Environmental Pollution 112 (2000) 1-9 horizontally in the three soil horizons (O, A, B). Three plots per site (0.5 and 8 km distance from the smelter) and four measurements per each plot were done (12 replicates in total). Because of the same soil type at the both locations (Derome and Lindroos, 1998), the soil measurements were done approximately from the same depths at both sites. The difference between the sites was, however, that the understorey vegetation is almost totally lacking at 0.5 km distance from the smelter (Mälkönen et a!., 1999). 2.5. Analysis of ABA For the ABA analyses, several current-year shoots of two E. nigrum patches were collected on 21 July, 18-19 and 26 August and 9 October at the two sites (0.5 and 8 km) simultaneously with other sampling. For methodo logical reasons the ABA contents in the stems collected on 21 July could not be analysed. The samples were fro zen in liquid nitrogen and stored at -80° C. The leaves and stems were freeze-dried, separated and ground with a mortar and pestle. One-hundred milligrams of homo genized plant material was suspended in 5 ml of 0.05 M phosphate buffer (pH 8.0), and 50 ng internal standard ([ 2 H 4]ABA) was added. After extraction in darkness, shaking at 4°C for a minimum of 2 h, the samples were centrifuged at 5000 rpm for 15 min. The supernatant was removed, and the pellet was resuspended in 2 ml of 0.05 mM phosphate buffer (pH 8.0) and centrifuged at 5000 rpm for 15 min. The supernatants were combined, the pH was adjusted to 2.7 with 0.1 M HCI, and partitio nated three times against equal volumes of ethyl acetate. The pooled ethyl acetate phase was reduced to dryness in vacuo in a rotary evaporator at 40° C. The residue was dissolved in 100 nl of 80% methanol and injected into a Radial Pak™ (Waters, Milford, USA) Cjg high-performance liquid chromatography (HPLC) column (Bxloo mm, 4 |im particle size). The HPLC system consisted of a Waters 510 pump, 600E control unit, 712 autosampler and 486 UV detector. ABA was eluted from the column by a linear gradient using a binary solvent system consisting of methanol and water, both containing 30 mM acetic acid, with a flow rate of 2ml min -1 . The gradient was from 30 to 50% methanol within 20 min. The fraction containing ABA was collected (15.5-17.5 min), and reduced to dryness in vacuo. The residue was dissolved in 100 (il of methanol and methylated in 500 nl of diazomethane dissolved in ether for 30 min. The sample was dried under a stream of N2 , dissolved in 12 nl of heptane, and 1 of this solution was injected into the gas chromatography-mass spec trometry (GC-MS). The GC-MS system consisted of a GCBO6O gas chromatograph with a A2OOS autosampler (Fisons, Milan, Italy), and a Platform mass spectro meter (Micromass, Altrincham, UK). The injector was operated at 230° C using the splitless mode and a CP SIL BCB Low Bleed MS (25 mx 0.25 mm i.d., 0.12 |im coating) column with a 2.5 m deactivated guard column (0.35 mm i.d., Chrompack Middelburg, The Nether lands) was used. Helium was used as the carrier gas at a flow rate of 25 cm s~'. The column temperature was programmed to change in a three slope linear gradient. The first phase was at 50° C for 2 min, then increased from 50 to 160° C at 15° C min -1 . The second slope was from 160 to 210° C at 3°C min -1 , and the third slope was from 210 to 280° C at 20° C min - '. The MS source was operated in electron impact mode at 70 eV at 180° C, and the interface at 250° C. lons of m/z 162, 190, 193 and 194 were monitored. 2.6. Statistical analysis To evaluate the effects of heavy metal deposition on E. nigrum the means of the measured parameters (soil moisture, chlorophyll, organic acid, ABA contents, stem water potential of plants) deriving from the two sites were compared by Mest when logarithmic transforma tions were used to normalise the data (Sokal and Rohlf, 1995; Statistix, 1996). Otherwise the Kruskal-Wallis comparison of mean ranks test was used (Sokal and Rohlf, 1995; Statistix, 1996). 3. Results 3.1. Chlorophyll contents Plant chlorophyll (a +b) contents were lower at 0.5 than at 8 km, and the differences between the means were statistically significant through the whole season (P < 0.05) (Fig. 1 a). The means of the chlorophyll (a + b) contents varied between 1.9 and 3.2 nmol chlorophyll g -1 dry wt. and were the lowest in October compared to the other sampling dates. The chlorophyll a/b ratio, was generally lower at a distance of 8 km than at 0.5 km, and the difference between the means at two distances was statistically significant (F <0.05) in mid August (Fig. lb). 3.2. Citric and malic acid contents of The citric acid contents in the leaves and stems of E. nigrum were higher at 8 km than at 0.5 km. In the stems, the difference between the means was statistically sig nificant (P< 0.05) (Fig. 2a). In the leaves, the mean pools of citric acid were 16 nmol mg -1 dry wt. at Bkm and 13 nmol mg -1 dry wt. at 0.5 km, and thus exceeded those in the stems of 14 nmol mg -1 dry wt. and 11 nmol mg - ' dry wt., respectively (Fig. 2a). The pools of malic acid in the leaves and stems of E. nigrum were higher at 8 km than at 0.5 km. The S. Monni et ai. / Environmental Pollution 112 (2000) 1-9 5 Fig. 1. (a) Chlorophyll (a+b) contents (µmol chlorophyll g-l dry weight) and (b) chlorophyll a/b ratios in the leaves of Empetrum nigrum at 0.5 and 8 km distances from the smelter in four collection dates. Bar indicates standard deviation (n = 5—10), and an asterisk indicates significant difference (P<0.05) between the means at two distances. difference between the means of the malic acid content of the stems was statistically significant (P <0.05; Fig. 2b). In the leaves, the mean malic acid values exceeded those in the stems, and were 17 nmol mg -1 dry wt. at Bkm and 14 nmol mg -1 dry wt. at 0.5 km compared to 13 and 8 nmol mg -1 dry wt., respectively, in the stems (Fig. 2b). 3.3. Stem water potential of E. nigrum and soil moisture The stem water potential of E. nigrum was more negative during the day (—l5 to —2l bars) than at night (-4 to —l2 bars; Fig. 3a, b). During the day, the water potential of plants at 8 km was more negative than those at 0.5 km. However, during the night the opposite was found (more negative at 0.5 km distance). The means of the stem water potential measured during the day and at night in July and during the day in August differed significantly between the plants growing at the two sites (P < 0.05) (Fig. 3a, b). The soil moisture content showed seasonal variation and decreased with depth (Fig. 4a, b). In the organic layer, soil moisture was higher at 8 km than at 0.5 km both during the day and at night in July. In August, however, it was higher at the site at 0.5 km. The differ ences between the means were statistically significant at night (P<0.05). In the mineral soil (A horizon) the trend was somewhat similar, and the difference between Fig. 2. (a) Citric acid and (b) malic acid contents (nmol mg -1 dry weight) in the leaves and stems of Empetrum nigrum collected at 0.5 and 8 km distances from the smelter. Bar indicates standard deviation (n = 5-6), and an asterisk indicates significant difference (,P<0.05) between the means at two distances. the means at the two sites was statistically significant during the day in August (P < 0.05) (Fig. 4a, b). In the lower mineral soil layer (B horizon), too, the trend was similar to that in the organic layer, and there were sig nificant differences between the means measured during the day (P<0.05) (Fig. 4a). At night the soil moisture content was the same at both sites (Fig. 4b). 3.4. ABA contents Empetrum plants growing at 0.5 km had higher con tents of ABA in their stems compared to those growing at 8 km (Fig. sa). With the exception of the July sam pling, a similar pattern was also observed in the leaves (Fig. Sb). No statistical differences were found between the means (P > 0.05). In late autumn the ABA contents decreased in the stems, but not in the leaves. The mean ABA contents in the leaves varied between 66-232 and 56-196 ng g~' dry wt., and in the stems between 45-113 and 33-71 ng g _l dry wt. at 0.5 and 8 km, respectively. 4. Discussion Chlorophyll contents in the leaves of E. nigrum were, in general, within the range reported for unpolluted 6 5. Monni et ai. I Environmental Pollution 112 (2000) 1-9 Fig. 3. Stem water potential of Empetrum nigrum (a) during the day and (b) at night at 0.5 and 8 km distances from the smelter. Bar indi cates standard deviation (n = 27-30), and an asterisk indicates sig nificant difference (P< 0.05) between the means at two distances. locations in Swedish Lapland for Empetrum hermaph roditum (Michelsen et al., 1996). However, at the 0.5 km site, the chlorophyll contents in E. nigrum were 15-30% lower than the values for plants growing at 8 km. Simi lar decreases in chlorophyll contents have been reported for several broadleaved species exposed to metals (Angelov et al., 1993; Bishnoi et al., 1993). The reduc tion in leaf chlorophyll contents may contribute to the stunted growth of E. nigrum near to the smelter. Shevt sova (1998) reported a 30% decrease in the length growth of terminal shoots of E. nigrum near Cu-Ni smelters in the Kola Peninsula compared to those from an unpolluted area. The growth of E. nigrum near to the Harjavalta smelter has also decreased (Salemaa et al., 1995). Near the smelter, the total Fe concentrations are increased and Mg concentrations decreased in E. nigrum tissue compared to further distance (Uhlig, unpublished results). Also, for example, pine is suffering from the Mg deficiency (Derome and Nieminen, 1998; Nieminen et al., 1999), and the pattern for exchangeable con centrations of Mg and Fe in the organic soil is similar to that of the plant parts (Derome and Lindroos, 1998). As being the constituents of chloroplasts (Marschner, 1995), the concentrations of these elements in E. nigrum Fig. 4. Soil moisture (%vol) of organic and two mineral soil layers (a) during the day and (b) at night at 0.5 and 8 km distances from the smelter. Bar indicates standard deviation (n=l2) and an asterisk indicates significant difference (P<0.05) between the means at two distances. J, July; A, August. and soil might partly explain the decreased contents of chlorophylls near the smelter. Chlorosis is the main symptom of Mg deficiency (Marschner, 1995) and excess Fe may catalyze formation of hydroxyl radicals in the chloroplasts being an early event of drought induced damage in photosynthetic tissue in dry condi tions (Price and Hendry, 1991). Citric acid contents in the leaves and stems of E. nigrum were generally lower than the values reported by Lee et al. (1978). These authors found citric acid con tents of up to 14-27 nmol mg -1 dry wt. in plant shoots (Homalium sp.) containing 100-1000 mg kg -1 Ni. Moreover, Ni hyperaccumulators (e.g. Sebertia sp., Hybanthus sp.) contained two times more citric acid than 'normal' plants (close to 70 nmol mg -1 dry wt.). However, there was much variation in citric acid con tents between species (Lee et al., 1978); levels up to 440 nmol mg -1 dry wt. citric acid have been reported in grass shoots exposed to Ni (Yang et al., 1997). The values reported in the literature for malic acid (malate) differ considerably depending on the plant species investigated (Mathys, 1977; Yang et al., 1997). In Cu- and Ni-resistant Silene cucubalus leaves, malate 5. Monni et al. j Environmental Pollution 112 (2000) 1-9 7 Fig. 5. Abscisic acid (ABA) contents (ng g- 1 dry weight) in (a) stems and (b) leaves of Empeirum nigrum at 0.5 and 8 km distances from the smelter in three to four collection dates. Bar indicates standard devia tion (n = 2). There was no statistical difference between the means (P> 0.05). content was up to 3 nmol mg" 1 fresh wt., when exposed to Zn (Mathys, 1977). This is about the same as values found here in the leaves and stems (8-17 nmol mg" 1 dry wt.) of E. nigrum. In grass species (maize, ryegrass), malic acid content increased to up to 375 nmol mg" 1 dry wt. after exposure to Ni (Yang et al., 1997). The organic acid contents in E. nigrum stems and leaves were lower in the plants growing at 0.5 km than at 8 km distance from the Cu-Ni smelter. In plants, elevated pools of organic acids are usually connected to elevated heavy metal levels in the substrate (especially Ni and Zn), and positive correlations have been found between organic acid and Ni and Zn concentrations in a number of studies (Lee et al., 1978; Godbold et a!., 1984; Yang et al., 1997). Organic acids are known to take part in the uptake and transport of metals, and accumulate in the cytosol or vacuoles of plants (Ernst, 1975; Ernst et al., 1992). Therefore, the concentrations of organic acids usually increase due to Ni and Zn stress (Lee et al., 1978; Godbold et al., 1984; Yang et al., 1997). However, in some studies a decline in organic acids under heavy metal stress has been reported (Foy et al., 1987). The results of our study do not follow the gen erally accepted pattern. At Harjavalta, Ni and Zn con centrations in the soil are elevated at 0.5 km from the smelter (Derome and Lindroos, 1998), where citric acid contents of II and !3 nmol mg~' dry wt. in the E. nigrum stems and leaves were found. At this site, Ni contents in the E. nigrum stems were between 63 and 164 mg kg -1 , and in the leaves 32-115 mg kg -1 (Uhlig, unpublished results). In plants growing at 8 km, the respective citric acid contents were 14 nmol mg" 1 dry wt. in the stems and 16 nmol mg -1 dry wt. in leaves. Moreover, in plants collected from a site at 4 km from the smelter, total Ni concentrations in leaves and stems of E. nigrum were 13-39 and 8-16 mg kg -1 , respectively (Uhlig, unpublished results). Therefore, it would appear that the total Ni concentrations in E. nigrum are close to background values at 8 km from the smelter. However, the lower metabolic efficiency meas ured by the decreased chlorophyll contents might explain the decreased production of organic acids near the smelter. The mean citric and malic acid contents in the leaves exceeded those in the stems of E. nigrum, although heavy metals are known to preferentially accumulate in the stems of E. nigrum (Uhlig et al., 1996; Monni et ai., 2000). The highest malate contents reported by Mathys (1977) occurred in either the leaves or stems, depending on the plant species. There were not large differences between the contents of malic and citric acids in both leaves and stems of E. nigrum. In the leaves of herbs and grasses either malic or citric acid contents were higher, depending on the plant species (Ernst, 1975) but in E. nigrum berries (Kallio and Markela, 1982), the malic acid contents exceeded citric acid contents. It has been speculated that drought resistance could help plants to survive in polluted areas (Pandolfini et al., 1996). E. nigrum is highly tolerant to drought and the anatomy of both its leaves (Wollenweber et al., 1992) and vascular tissue is xeromorphic (Carlquist, 1989). ABA .contents in E. nigrum stems and leaves were partly in agreement with results of earlier studies that showed heavy metal-induced ABA accumulation (Rauser and Dumbroff, 1981; Poschenrieder et al., 1989). Plant ABA contents were higher near to the smelter in August and October, although no significant differences were found due to the low number of replicates. The reasons for the high ABA contents in the leaves in October are not known, but they may be related to environmental con ditions, e.g. decreasing temperature, day length and ageing of the leaves. The stem water potential and leaf ABA contents at both sites correlated in July, but not in August, accord ing to an overall pattern, e.g. the ABA content in the leaves usually rises as the leaf water potential falls (Beardsell and Cohen, 1975). However, in the early stages of soil desiccation, ABA is transported as a che mical signal from the roots to the leaves where, by enhancing stomatal closure, it prevents a decline in 8 S. Monni et ai. / Environmental Pollution 112 (2000) 1-9 water potential (Davies and Zhang, 1991). Additionally, in nutrient-deficient plants, production of ABA seems to occur at less negative water potentials (Radin, 1984). Near to the smelter at Harjavalta, E. nigrum is suffering from decreased concentrations of Mg and Mn, but other nutrients (N, P, Ca, K) in the plant tissue are not significantly lower near the smelter (Uhlig, unpublished results). In addition to nutrient deficiency, plants might require less water for growth and photosynthesis as these metabolic processes have declined. The roots, where the ABA synthesis occurs (Marschner, 1995), are damaged near the smelter, which could in turn affect the ABA synthesis and measured ABA contents. The soil moisture content was almost the same during day and night. The soil moisture differed from the stem water potential measurements at night, the more nega tive stem water potential at night near the smelter cor relating in general with the soil moisture in July but not in August. The stem water potential at night is close to the soil water potential because the stomata are closed (Aber and Melillo, 1991). However, the prevailing weather conditions and heterogeneity of the soil are factors that might affect the results. The impaired water holding capacity at 0.5 km (Derome and Nieminen, 1998), would support the more negative stem water potential at night, close to the smelter. 5. Conclusions Our results indicate that heavy metal pollution has negative effects on E. nigrum. The decreased contents of chlorophyll pigments and organic acids and increased ABA contents indicate a reduction in the physiological activity of E. nigrum near to the pollution source. Although E. nigrum is known to be one of the most tolerant species near Cu-Ni smelters in the northern hemisphere (Helmisaari et ai., 1995; Uhlig et ai., 1996) and to accumulate high concentrations of Cu and Ni in its living parts in greenhouse experiments (Monni et ai., 2000), the emissions have clearly decreased the vitality of E. nigrum. Acknowledgements We would like to thank the staff of the Finnish Forest Research Institute, the Department of Plant Physiology and Microbiology, University of Tromso, and the Department of Physiological Ecology of Plants, Uni versity of Tubingen, for helping with the practical aspects of this study. Olavi Junttila and Maija Salemaa made critical comments on the manuscript and John Derome revised the English. The work was funded by Maj and Tor Nessling Foundation and NorFA. References Aber, J.D., Melillo, J.M., 1991. Terrestrial Ecosystems. Saunders College Publishing, Rinehart and Winston, Inc, Orlando, FL. Amiro, 8.D., Courtin, G.M., 1981. Patterns of vegetation in the vici nity of an industrially disturbed ecosystem, Sudbury, Ontario Canadian Journal of Botany 59, 1623-1639. Angelov, M., Tsonev, T., Uzunova, A., Gaidardjieva, K., 1993. Cu 2 + effect upon photosynthesis, chloroplast structure, RNA and protein synthesis of pea plants. Photosynthetica 28, 341-350. Antonovics, J., Bradshaw, A.D., Turner, R.G., 1971. Heavy metal tolerance in plants. Advances in Ecological Research 7, 1-85. Bagatto, G., Shorthouse, J.D., 1991. Accumulation of copper and nickel in plant tissues and an insect gall of lowbush blueberry, Vac cinium angustifolium, near an ore smelter at Sudbury, Ontario, Canada. Canadian Journal of Botany 69, 1483-1490. Baker, A.J.M., 1981. Accumulators and excluders strategies in the response of plants to heavy metals. Journal of Plant Nutrition 3, 643-654. Baker, A.J.M., 1987. Metal tolerance. New Phytologist 106, 93-111 Beardsell, M.F., Cohen, D., 1975. Relationships between leaf water status, abscisic acid levels and stomatal resistance in maize and sor ghum. Plant Physiology 56, 207-212. Becerril, J.M., Conzälez-Murua, C., Munoz-Rueda, A., De Felipe, M.R., 1989. Changes induced by cadmium and lead in gas exchange and water relations of clover and lucerne. Plant Physiology and Biochemistry 27, 913-918. Bell, J.N.8., Tallis, J.H., 1973. Biological flora of the British Isles. Empetrum nigrum L. Journal of Ecology 61, 289-305. Bishnoi, N.R., Sheoran, 1.5., Singh, R., 1993. Influence of cad mium and nickel on photosynthesis and water relations in wheat leaves of different insertion level. Photosynthetica 28, 473-479. Boehringer, 1989. Methods of Biochemical Analysis and Food Analy sis Using Single Reagents. Boehringer Mannheim GMBH, Bio chemica, Mannheim, Germany. Bradley, R., Burt, A.J., Read, D.J., 1981. Mycorrhizal infection and resistance to heavy metal toxicity in Calluna vulgaris. Nature 292, 335-337. Bradley, R., Burt, A.J., Read, D.J., 1982. The biology of mycorrhiza in the Ericaceae. VIII. The role of mycorrhizal infection in heavy metal resistance. New Phytologist 91, 197-209. Bringezu, K., Lichtenberger, 0., Leopold, 1., Neumann, D., 1999. Heavy metal tolerance of Silene vulgaris. Journal of Plant Physiol ogy 154, 536-546. Carlquist, S., 1989. Wood and bark anatomy of Empetraceae; com ments on paedomorphosis in woods of certain small shrubs. Aliso 12, 497-515. Davies, W.J., Zhang, J., 1991. Root signals and the regulation of growth and development of plants in drying soil. Annual Review of Plant Physiology and Plant Molecular Biology 42, 55-76. Derome, J., Lindroos, A.-J., 1998. Effects of heavy metal contamina tion on macronutrient availability and acidification parameters in forest soil in the vicinity of the Harjavalta Cu-Ni smelter, SW Fin land. Environmental Pollution 99, 225-232. Derome, J., Nieminen, T., 1998. Metal and macronutrient fluxes in heavy-metal polluted Scots pine ecosystems in SW Finland. Envir onmental Pollution 103, 219-228. Elvebakk, A., Spjelkavik, S., 1995. The ecology and distribution of Empetrum nigrum ssp. hermaphroditum on Svalbard and Jan Mayen. Nordic Journal of Botany 15, 541-552. Ernst, W.H.0., 1975. Physiology of heavy metal resistance in plants. Proceedings of the International Conference of Heavy Metals in the Environment, Toronto 2, 121-136. Ernst, W.H.0., Verkleij, J.A.C., Schat, H., 1992. Metal tolerance in plants. Acta Botanica Neerlandica 41, 229-248. S. Monni et ai. / Environmental Pollution 112 (2000) 1-9 9 Foy, C.D., Lee, E.H., Wilding, 5.8., 1987. Differential aluminium tolerances of two barley cultivars related to organic acids in their roots. Journal of Plant Nutrition 10, 1089-1101. Godbold, D.L., Horst, W.J., Collins, J.C., Thurman, D.A., Marsch ner, H., 1984. Accumulation of zinc and organic acids in roots of zinc tolerant and non-tolerant ecotypes of Deschampsia caespitosa. Journal of Plant Physiology 116, 59-69. Goldbach, E., Goldbach, H., Wagner, H., Michael, G., 1975. Influ ence of N-deficiency on the abscisic acid content of sunflower plants. Physiologia Plantarum 34, 138-140. Graan, T., Ort, D.R., 1984. Quantitation of the rapid electron donors to P7OO. the functional plastoquinone pool, and the ratio of the photosystems in spinach chloroplasts. The Journal of Biological Chemistry 259, 14003-14010. Hampp, R., Goller, M., Fullgraf, H., 1984. Determination of com partmented metabolite pools by a combination of rapid fractiona tion of oat mesophyll protoplasts and enzymic cycling. Plant Physiology 75, 1017-1021. Helmisaari, H.-S., Derome, J., Fritze, H., Nieminen, T., Palmgren, K., Salemaa, M., Vanha-Majamaa, 1., 1995. Copper in Scots pine for ests around a heavy-metal smelter in South-Westem Finland. Water, Air and Soil Pollution 85, 1727-1732. Hogan, G.D., Rauser, W.E., 1979. Tolerance and toxicity of cobalt, copper, nickel and zinc in clones of Agrostis gigantes. New Phytol ogist 83, 665-670. Kallio, H., Markela, E., 1982. Nonvolatile acids in the juice of crow berry, Empetrum nigrum coll. Journal of Food Science 47, 1466 1468. Laaksovirta, K., Silvola, J., 1975. Effect of air pollution by copper, sulphuric acid and fertilizer factories on plants at Harjavalta, W. Finland. Annales Botanici Fennici 12, 81-88. Lamoreux, R.J., Chaney, W.R., 1978. The effect of cadmium on net photosythesis, transpiration, and dark respiration of excised silver maple leaves. Physiologia Plantarum 43, 231-236. Lee, J., Reeves, R.D., Brooks, R.R., Jaffre, T., 1978. The relation between nickel and citric acid in some nickel-accumulating plants. Phytochemistry 17, 1033-1035. Mälkönen, E., Derome, J., Fritze, H., Helmisaari, H.-S., Kukkola, M., Kytö, M., Saarsalmi, A., Salemaa, M., 1999. Compensatory fertili zation of Scots pine stands polluted by heavy metals. Nutrient Cycling in Agroecosystems 55, 239-268. Marschner, H., 1995. Mineral Nutrition of Higher Plants, 2nd Edi tion. Academic Press, London. Mathys, W., 1977. The role of malate, oxalate and mustard oil gluco sides in the evolution of zinc-resistance in herbage plants. Physiolo gia Plantarum 40, 130-136. Michelsen, A., Jonasson, S., Sleep, D., Havström, M., Callaghan, T.V., 1996. Shoot biomass, 5 13 C, nitrogen and chlorophyll respon ses of two arctic dwarf shrubs to in situ shading, nutrient application and warming simulating climatic change. Oecologia 105, 1-12. Monni, S., Salemaa, M., Millar, N., 2000. The tolerance of Empetrum nigrum to copper and nickel. Environmental Pollution 109, 221-229. Nieminen, T.M., Derome, J., Helmisaari, H.-S., 1999. Interactions between precipitation and Scots pine canopies along a heavy-metal pollution gradient. Environmental Pollution 105, 129-137. Ojansuu, R., Henttonen, H., 1983. Kuukauden keskilämpötilan, läm pösumman ja sademäärän paikallisten arvojen johtaminen ilmatie teen laitoksen mittaustiedoista. [English summary: Estimation of local values of monthly mean temperature, effective temperature sum and precipitation sum from the measurements made by the Finnish meteorological office.] Silva Fennica 17, 143-160. Pandolfini, T., Gabbrielli, R., Ciscato, M., 1996. Nickel toxicity in two Durum wheat cultivars differing in drought sensitivity. Journal of Plant Nutrition 19, 1611-1627. Poschenrieder, C., Gunse, 8., Barcelo, J., 1989. Influence of cadmium on water relations, stomatal resistance, and abscisic acid content in expanding bean leaves. Plant Physiology 90, 1365-1371. Price, A.H., Hendry, G.A.F., 1991. Iron-catalysed oxygen radical for mation and its possible contribution to drought damage in nine native grasses and three cereals. Plant, Cell and Environment 14, 477-484. Radin, J.W., 1984. Stomatal responses to water stress and to abscisic acid in phosphorus-deficient cotton plants. Plant Physiology 76, 392-394. Rantalahti, R., 1995. Outokummun Harjavallan tehtaiden ympär istönsuojelusta. Vaskenviesti 3-4, 2-5. Rauser, W.E., 1984. Partial purification and characterization of cop per-binding protein from roots of Agrostis gigantea Roth. Journal of Plant Physiology 115, 143-152. Rauser, W.E., Curvetto, N.R., 1980. Metallothionein occurs in roots of Agrostis tolerant to excess copper. Nature 287, 563-564. Rauser, W.E., Dumbroff, E. 8., 1981. Effects of excess cobalt, nickel and zinc on the water relations of Phaseolus vulgaris. Environmental and Experimental Botany 21, 249-255. Reilly, C., 1969. The uptake and accumulation of copper by Becium homblei (De Wild. ) Duvig. & Plancke. New Phytologist 68, 1081— 1087. Ritchie, G.A., Hinckley, T.M., 1975. The pressure chamber as an instrument for ecological research. In: Macfadyen, A. (Ed.), Advances in Ecological Research 9, pp. 165-254. Salemaa, M., Ojala, A., Vanha-Majamaa, 1., Vieno, M., 1995. Life history of the crowberry Empetrum nigrum L. in heavy metal pol luted environment. In: Oborny, 8., De Kroon, H., van Groe nendael, J. (Eds.), The Role of Clonality in Plant Communities. Abstracts of the 4th International Workshop on Clonal Plants. Visegard, Hungary, 18-22 April, p. 46. Scholander, P.F., Hammel, H.T., Bradstreet, E.D., Hemmingsen, E.A., 1965. Sap pressure in vascular plants. Science 148, 339-346. Shevtsova, A., 1998. Responses of subarctic dwarf shrubs to climate change and air pollution. Annales Universitatis Turkuensis Ser. All, 113. Doctoral thesis. Shorthouse, J.D., Bagatto, G., 1995. Potential role of lowbush blue berry (Vaccinium angustifolium) in colonizing metal-contaminated ecosystems. In: Gunn, J.M. (Ed.), Restoration and Recovery of an Industrial Region. Progress in Restoring the Smelter-Damaged Landscape near Sudbury, Canada. Springer-Verlag, New York, pp. 247-255. Sokal, R.R., Rohlf, F.J., 1995. Biometry. W.H. Freeman, New York. Statistix, 1996. Statistix for Windows. User's Manual. Borland Inter national Inc, Tallahassee, FL. Uhlig, C., Salemaa, M., Vanha-Majamaa, 1., 1996. Element distribu tion in Empetrum nigrum microsites at heavy metal contaminated sites in Harjavalta, Western Finland. In: Kopponen, P., Kär enlampi, S., Rekilä, R., Kärenlampi, L. (Eds.), Bio- and Eco technological Methods in Restoration. Abstracts of an International Advanced Course and Minisymposium, 16-18 December, Kuopio, pp. 21. Wollenweber, E., Dorr, M., 1992. Stelzer, R, Arriaga-Giner, F. J., Lipophilic phenolics from the leaves of Empetrum nigrum che mical structures and exudate localization. Botanica Acta 105, 300- 305. Woolhouse, H.W., 1983. Toxicity and tolerance in the responses of plants to metals. In: Lange, 0.L., Nobel, P.S., Osmond, C.8., Zieg ler, H. (Eds.), Encyclopedia of Plant Physiology. Springer-Verlag, Berlin, pp. 245-300. Yang, X.E., Baligar, V.C., Foster, J.C., Martens, D.C., 1997. Accu mulation and transport of nickel in relation to organic acids in rye grass and maize grown with different nickel levels. Plant and Soil 196, 271-276. Monni, S., Bucking, H., Kottke, 1., 2000. Ultrastructural element localization by EDXS in Empetrum nigrum. Manuscript. V ULTRASTRUCTURAL ELEMENT LOCALIZATION BY EDXS IN EMPETRUM NIGRUM l,2 Satu Monni . 3 Heike Bucking, 'ingrid Kottke, 'Vantaa Research Centre. Finnish Forest Research Institute. P.O. Box 18. FIN-01301. Vantaa, Finland "Department of Ecology and Systematics, University of Helsinki, P.O. Box 7. FIN-00014 Helsinki, Finland 3 Centre for Environmental Research and Technology, Applied Botany, Plant Anatomy and Plant Physiology, University of Bremen, D-28359, Bremen, Germany department of Special Botany and Mycology, University of Tubingen, D-72076, Tubingen, Germany Abstract E. nigrum is one of the few species growing on highly polluted areas in the northern boreal forests and accumulates considerable amounts of heavy metals especially in its older stems. Previous-year stems of Empetrum nigrum were collected from two different sites located at distances of 0.5 km (highly contaminated) and 8 km (low contaminated) from a Cu-Ni smelter at Harjavalta, SW Finland. The element (Al. As. Cu, Fe, Mn, Zn, Ca, K. P. S. Mg. Na) localization was performed by energy-dispersive X-ray spectroscopy (EDXS) after crvofixation. freeze-drving and pressure infiltration of the material. The results showed higher amounts of Cu. As and Fe in cell compartments of E. nigrum close to the smelter than in further distance. The Al and Zn amounts, in contrast, showed no clear differences between the sites. Cu was distributed homogeneously in the tissue and occurred in vacuoles, cytoplasm, cell walls as well as in lumens of the vascular tissue. The higher amounts of As were localized in the primary cell walls of living (ray cells, phloem) than dead cells (xylem, sclereids). but also vacuoles and cytoplasm contained elevated amounts of As. Ray cells, phloem and sclereids had elevated Fe amounts compared to the other tissues in the contaminated stem samples but owing to the high variation between the replicates, no significant differences were found. Also Fe was localized in the cell walls, cytoplasm and * Corresponding author, address: Vantaa Research Centre. Finnish Forest Research Institute. Box 18. FIN-01301 Vantaa. Finland; e-mail: Satu.Monni@metla.fi; fax: +358-9- 85705569 2 vacuoles. Based on the rather homogeneous localization of Cu. As and Fe in the living tissue and increased amounts of Cu. As and Fe in vacuoles, cell walls and cytoplasm near the smelter, it seems that not only one specific mechanism contribute to the heavy metal tolerance of E. nigrum. The role of complexing agents in heavy metal tolerance in the cytoplasm or vacuoles could not be shown by this study. Because of the more frequent localization of electron dense phenolic material in the polluted samples, it might also have a function in the heavy metal tolerance of E. nigrum. Key words Empetrum nigrum L., crowberry, heavy metal localization, nutrient localization. Cu, As, Fe. Ca. K. tolerance, transmission electron microscopy. EDXS 1. Introduction The heavy metal tolerance of plants is based on different biochemical mechanisms and is often species and metal specific (e.g. Antonovics et al., 1971; Baker. 1981; Woolhouse. 1983; Baker and Walker, 1990). Plants can produce intracellular metal-chelating substances, so that the intracellular availability of metals is maintained within certain limits (Verkleij and Schat. 1990). The accumulation of metal-chelating substances upon exposure to excessive metal amounts has been found. Organic acids (malic and citric acid) are important in chelating Ni and Zn (Lee et al, 1978; Yang et al., 1997), while Cu has been found to be associated with proteins and phenolic compounds in leaves and roots (Neumann et al., 1995). Deleterious amounts of metals can also be translocated and stored in certain cell organelles, where sensitive metabolic activities do not take place (Verkleij and Schat. 1990). The translocation of metals varies, but generally cell walls, vacuoles, intercellular spaces and cytoplasm of roots and leaves contain elevated metal (Cu. Fe, Zn. Pb) amounts (Mullins et al., 1985; Neumann et al, 1995; Neumann et al, 1997; Lichtenberger and Neumann. 1997). For example in leaves. Ni has been found to be accumulated in the chloroplasts of the bundle sheath cells (L'Huillier et al. 1996). The ecology (Bell and Tallis, 1973) and wood anatomy (Miller, 1975; Carlquist, 1989) of crowberrv (Empetrum nigrum L.) have been extensively studied, but its heavy metal tolerance mechanism on the cellular or subcellular level is not well known. Besides wide 3 ecological amplitude (Bell and Tallis. 1973), E. nigrum grows on highly heavy metal contaminated sites in the vicinity of Cu-Ni smelters in the northern hemisphere (Laaksovirta and Silvola. 1975; Chertov et al.. 1993; Helmisaari et al., 1995; Uhlig et al.. 2000). It grows also in serpentine soils (Proctor and Woodcll. 1971), which typically contain elevated amounts of Ni, Cr and Mg and low levels of Ca (Proctor, 1971). E. nigrum has been found to accumulate relatively high concentrations of Cu and Ni in greenhouse and field especially in older plant parts (Helmisaari et al., 1995; Monni et al., 2000 a; Uhlig et al., 2000), whereas the transport of metals to the green leaves is restricted (Monni et al., 2000 a). The structure of the vascular tissue of the stem wood is very xeromorphic; the large number of vessels and the presence of tracheids of the imperforate tracheary element type lead to a high conductive safety of the plant (Miller, 1975; Carlquist, 1989). The use of transmission electron microscopy (TEM) equipped with electron analyzers has made the ultrastructural localization of elements in plants possible. Heavy metal tolerance mechanisms and the localization of elements in several plant species, which contain relatively high amounts of metals in their tissues, have been studied by electron dispersive X-ray spectroscopy (EDXS) and electron energy loss spectroscopy (EELS) (Mullins et al.. 1985; Turnau et al., 1993 a. b; Neumann et al., 1995; Lichtenberger and Neumann. 1997). However, both methods have their advantages and limitations (Stelzer and Lehmann. 1993; Kottke. 1994; Bucking et al.. 1998). The aim of the current X-ray microanalytical study was to investigate heavy metal and nutrient localization in stems of E. nigrum from a highly and a moderately heavy metal polluted area in order to obtain more information about possible tolerance mechanisms of E. nigrum. Previous-year stems were selected for the experiments, because the highest concentrations of metals have been measured in older tissues, especially in stems (Monni et al., 2000 a; Uhlig et ai.. 2000). 2. Materials and methods 2.1. The Harjavalta smelter area The previous-year stems were collected at a distance of 0.5 (high contamination) and Bkm (low contamination) to the south-east from the Cu-Ni smelter at Harjavalta. SW Finland. 4 The Cu smelter was established in 1945 and the Ni smelter in 1960. Sulphur dioxide and heavy metals have been emitted into the environment for the past 40-60 years. The deposition of metals near the smelter was considerably reduced in the 1990' s after a taller stack was built and electrostatic filters were installed (Rantalahti, 1995). The prevailing wind direction has been from the south, south-west and south-east (Derome, 2000). The sites were chosen to the south-east of the smelter because it was the only direction with two sites representing the same forest site type (Calluna site type) and soil type (orthic podzol). The pH of the organic layer was 3.5 at 0.5 km and 3.6 at 8 km distance from the smelter (Derome and Lindroos, 1998). The site at 0.5 km is located in a heavily polluted area where the total element concentrations in the organic layer are 5 800 mg Cu kg" 1 dw, 460 mg Ni kg" 1 dw. 18 600 mg Fe kg" 1 dw. 520 mg Zn kg" 1 dw. 1 560 mg Al kg" 1 dw. 30 mg Mil kg" 1 dw, 970 mg Ca kg" 1 dw, 380 mg K kg" 1 dw and 440 mg Mg kg" 1 dw. Although the background values are not reached, the soil is only slightly contaminated at a distance of 8 km from the smelter, the respective concentrations being 150 mg Cu kg" 1 dw, 40 mg Ni kg" 1 dw. 2 200 mg Fe kg" 1 dw, 60 mg Zn kg" 1 dw, 1 760 mg Al kg" 1 dw, 60 mg Mn kg" 1 dw, 970 mg Ca kg" 1 dw, 400 mg K kg" 1 dw and 210 mg Mg kg" 1 dw. Near the smelter Ca, Mg and K cations are displaced from cation exchange sites by Cu and Ni cations in the organic layer, the exchangeable concentrations of those nutrients being much lower near the smelter than further distance (Derome and Lindroos. 1998). 2.2. Preparation technique for EDXS analysis Previous-year stems were collected in the field, cryofixed by liquid argon gas surrounded by liquid nitrogen and stored in liquid nitrogen (-192 °C). After cryofixation the stem pieces were freeze-dried (CFD, Leica. Germany) for 14 days under high vacuum (< lx 10" 5 mbar) and low temperature conditions (-100 °C) to avoid ice recrystallization in the cytoplasm of the cells (recrystallization temperature: -80 °C. Robinson et al., 1985). After freeze-drying, the stem samples were pressure infiltrated directly in 100 % Spurr's epoxy resin (Spurr, 1969), using a method described by Fritz (1980). Because some of the samples could not be infiltrated directly in 100% Spurr's epoxy resin, some intermediary steps using ether were added to the embedding protocol. For EDXS the embedded samples were dry sectioned (0.5 |im), placed on filmed Ni or Cu grids and carbon coated. 5 2.3. EDXS analysis The energy dispersive X-ray microanalytical studies were carried out under standardized conditions using a Philips EM 420 provided with the ED AX DX-4 system. EDXS spectra were collected between 0 and 20 keV with a Si(Li) X-ray detector (size: 10 mm 2 ) equipped with a thin beryllium window. An acceleration voltage of 120 kV, an objective aperture of 70 jun and a detection time of 100 live seconds were used. The calculated effective spot size (D eff) of the measurement points was 12 nm. However, based on the interactions of the electron beam with the specimen, the real spot size was slightly larger. One spectrum between 0.6 and 8.6 keV is shown in Fig. 1. The peak centre of the Ka-line of the individual elements was 1.04 keV for Na. 1.25 keV for Mg, 1.49 keV for Al, 1.74 keV for Si, 2.01 keV for P, 2.31 keV for S. 2.62 keV for CI. 3.31 keV for K, 3.69 keV for Ca, 5.89 keV for Mn. 6.40 keV for Fe, 7.45 keV for Ni. 8.04 keV for Cu, 8.63 keV for Zn and 10.53 keV for As. The element distribution was measured as a peak to background ratio (P/B) in order to minimize the effects of surface irregularities of the sections during analysis (Fig. 3). For the X-ray map a magnification of 6400 x, a resolution of 128 x 100 measurement points and a dwell time of 500 ms in the live second mode were used. The image was captured with 1024 x 800 points. Figure 1. Typical EDX spectra (0.6 - 8.6 keV) of the electron-dense material in the ray cell of an E. nigrum stem collected at 0.5 km distance from the Cu-Ni smelter. The CI peak is an artefact caused by the epoxy resin, Ni by the Ni-grid and Si by the dry-sectioning of glass knives and the Si(Li) X-ray detector system. EDXS point measurements were carried out in xylem, phloem, ray cells and sclereids (Fig. 2) of the stems. The lumen, primary and secondary cell wall of vessels, tracheids and 6 sclereids, as well as cytoplasm, primary cell wall and vacuole of ray cells, phloem and parenchyma of the primary ray, were analysed. Additional electron dense material found in the lumen of xylem vessels, vacuoles of ray cells and phloem were measured in both high and low contaminated samples. 2.4. Statistical analysis Altogether, four samples from both distances were studied. The measurement points per sample per tissue were generally around 10 (in some tissues between 1-17). To compare the distribution of elements in the different plant tissues at two distances, the Kruskal-Wallis test, a comparison of the mean ranks, was performed, because the logarithmic transformation did not normalize the data (Sokal and Rohlf, 1995; Statistix, 1996). Figure 2. Cross-section of a previous-year stem of E. nigrum collected at 0.5 km distance from the Cu-Ni smelter. Magnification 50 *. Scale bar 110 pm. PX = primary xylem, V = vessels, T = tracheids, R = ray cells, SX = secondary xylem, PH = phloem, PR = parenchyma of the primary ray, SC = sclereids. 7 3. Results 3.1. Localization of heavy metals in different tissues and cell compartments The heavy metal localization in stems of E. nigrum analysed by EDXS varied according to the metal and tissue, and the amounts of Cu, As and Fe were higher near to the smelter than in the stems from the control site (3a, b, c). In nearly all of the analysed tissues, the Cu amount was higher in the samples from the high contamination plots compared to the low contaminated stem samples (Fig. 3a). However, the difference was statistically significant only in some cell compartments (p < 0.05). The highest amounts were measured in the vessel lumens and primary wall of the ray cells. The Cu amounts were elevated both in living (ray cells, phloem, parenchyma of the primary ray) and in dead cells (xylem, sclereids), but no clear differences were found between the Cu amounts in different cell compartments (Fig. 3a). According to the X-ray microanalytical mappings, Cu was rather homogeneously distributed among the tissue (Fig. sc). The As amounts were statistically higher at 0.5 km distance in almost all cell compartments (p < 0.05) (Fig. 3b). The highest As amounts were measured in outer regions of the stem cross-section, in different cell compartments of the phloem, in vacuoles of the primary ray and in lumen and secondary wall of sclereids. Lower As amounts were generally detected in vessels and tracheids of the xylem (Fig. 3b). In general, higher Fe amounts were found in stems from the contaminated sites (Fig. 3c). The Fe amounts of the stems from the low contaminated site were under or close to the X ray microanalytical limit of detection. Owing to the high variation between the replicates, the differences were statistically significant only in the secondary wall of vessels, primary cell walls and vacuoles of ray cells, vacuoles of phloem, cytoplasm of the parenchyma of the primary ray and lumen of sclereids (p < 0.05). In general, ray cells, phloem and sclereids had higher Fe amounts than other tissues from the contaminated stem samples (Fig. 3c). However, due to the high standard error of mean generally no significant differences were found. 8 Figure 3a) - i). Peak to background (P/B) ratios of a) Cu, b) As, c) Fe, d) AI, e) Mn, f) Ca, g) K, h) P and i) S in lumen (L), secondary cell wall (SW), primary cell wall (PW), cytoplasm (C), vacuole (V) and electron-dense material (ED) of vessel, tracheid, ray cells, phloem, parenchyma of the primary ray, sclereids and lumen of dead cells (L) of E. nigrum previous year stems sampled at 0.5 and 8 km distances from the Cu-Ni smelter. 9 Fig. 3 (continued). The number of measurement points were: lumen (n = 18-21), secondary wall (n = 18-20) and primary wall of vessel (n = 5-10); lumen (n = 18-20), secondary wall (n = 18-20) and primary wall of tracheid (n = 3-5); primary wall (n = 11-18), cytoplasm (n = 10- 18), vacuole (n = 13-18) and electron dense material of ray cells (n = 1-6); primary wall (n = 5-13), cytoplasm (n = 10), vacuole (n = 11-13) and electron dense material of phloem (n = 5); 10 Fig. 3 (continued). Primary wall (n = 10-13), cytoplasm (n = 10-13) and vacuole (n = 10-13) of parenchyma of primary ray; lumen (n = 10-13), secondary wall (n = 10-13) and primary wall of sclereids (n = 10-12); electron dense material of lumen (n = 1-22). Bar indicates standard error of the mean (SE). Note the different scales. 11 The variation in the A 1 amounts between the different measurement points was very small (Fig. 3d; 4b). Although the A 1 amounts in many cell compartments were higher in the stems from the contaminated site, the difference between the two distances was very small, and the increased contents could not clearly be explained by an effect of the site. The highest A 1 amounts were detected in the electron-dense material of the ray cells, where particles were found with very high A 1 amounts (Fig. 3d. Sb). The Zn amounts were generally near to the detection limit (P/B ratio generally below 0.25) and Zn could only clearly be detected in living cells of the phloem and the primary ray tissue in stems from the highly contaminated site. In contrast to the elements described above, the Mn amounts in the different cell compartments of the stems were lower near to the smelter (Fig. 3e). In stems from the low contaminated site Mn was especially localized in living tissue (ray cells, phloem, primary ray cells) and the cell walls of sclereids. In contrast, in the xvlem of the vascular tissue only low Mn contents were detected. In stems from the highly contaminated site, the Mn amounts were close to the detection limit. In these samples no clear differences in the Mn amounts between the different cell compartments or tissues were found (Fig. 3e). 3.2. Localization of macronutrients and Na in different tissues and cell compartments The amounts of macronutrients varied between different cell compartments. The Ca amounts were generally higher at 8 km than at 0.5 km distance from the smelter, the highest amounts occurring in the electron-dense material of vessel lumens (Fig. 3f). Ca was mainly located in living parts of the tissue, in the phloem and primary ray and here especially in primary cell walls. The amounts in the primary cell walls of the dead tissue (vessels, tracheids. sclereids) were lower (Fig. 3f). The X-rav microanalvtical mapping showed that Ca was mainly located in primary cell walls and here especially in the pits between two cells (Fig. 4c). However no clear statistical differences (p < 0.05) were found in the Ca amounts between primary and secondary walls. The cytoplasm of the living tissue also contained relatively high amounts of Ca, whereas the amounts in lumens and vacuoles were low in both the high and low contaminated samples (Fig. 3f). 12 Figure 4. Vessel elements of E. nigrum stem at 8 km distance from the Cu-Ni smelter, a) vessel elements, b) aluminium in the lumen, c) calcium in the plasmodesmata and primary cell wall and d) potassium in the secondary cell wall of vessel elements. Magnification 6 400 x. Scale bar 1 pm. V = vessels, T = tracheids, PW = primary cell wall, SW = secondary cell wall, P = pit. The K amounts were higher at 0.5 km than at 8 km (Fig. 3g). The highest K amounts were measured mainly in living parts of the tissue, i.e. in phloem and primary ray cells. The primary walls of the phloem and parenchyma of the primary ray had higher K amounts than the primary walls of dead cells (vessels, tracheids, sclereids) (Fig. 3g). In contrast to Ca. relatively high K amounts were measured in primary and secondary walls of the xylem (Figs 3g, 4d) and sclereids (Fig 3g). The cytoplasm also had relatively high K amounts (Fig. sd), whereas only low amounts were found in lumens and vacuoles of both the high and low contaminated samples (Fig 3g). In the ray cells, there were higher K amounts in the electron-dense material and the cytoplasm than in the vacuoles and primary cell walls in the highly contaminated samples (Figs 3g, sd). 13 Figure 5. Vessel elements and ray cells of E. nigrum stem at 0.5 km distance from the Cu-Ni smelter, a) Vessel elements, b) aluminium and c) copper in the lumen of vessel elements, d) potassium, e) phosphorus and f) sulphur in the electron-dense material of vacuoles and cytoplasm of raycells. Magnification 6 400 *. Scale bar 1 pm. V = vessels, T = tracheids, R = ray cells, SW = secondary cell wall, P = pit. The P and S amounts did not vary according to the site (Figs 3h, i). Relatively high P and S amounts were found in ray cells, phloem and primary ray cells, whereas in dead cells, vessels and tracheids of the xvlem and the sclereids lower amounts were detected (Fig. 3h, i). In the living tissues, P and S were mainly located in the cytoplasm or the electron-dense 14 material (Fig. 3h, i. se. f). High P and S amounts were also found in the primary wall of the phloem and parenchyma of the primary ray whereas vacuoles contained lower amounts of both elements (Fig. 3h, i). The Mg and Na amounts were close to the X-ray microanalytical limit of detection, the peak to background ratio (P/B ratio) for Mg being generally below 0.4, and for Na below 0.1. The differences in the amounts were therefore most probably due to the background variation. 4. Discussion The heavy metal pollution had an influence on the cellular amounts of Cu, As and Fe in the E. nigrum stems. The results were in good agreement with earlier results obtained near to the Harjavalta Cu-Ni smelter. At 0.5 km from the Cu-Ni smelter, the Cu and Fe concentrations in the soil (Derome and Lindroos. 1998) and in E. nigrum (Helmisaari et ai, 1995; Uhlig et ah, 2000) are much higher than at e.g. 2. 4 or 8 km distance from the smelter. Although As concentrations in the soil and plants have not been earlier reported, the emissions from the smelter do in fact also contain As (Rantalahti, 1995). The Zn concentrations in the soil and plant parts are also elevated (Uhlig et al., 2000; Derome and Lindroos, 1998; Derome. 2000). The EDXS analyses did not indicate any clear differences in the Zn concentrations between the different sites, but this might be due to the detection limit of the EDXS and the difficulty to seperate the Zn peak when Cu was present in high amounts. The Mn amounts, however, were clearly higher in different cell compartments at 8 km than at 0.5 km. This supports the results of Derome and Lindroos (1998) and Uhlig et ah (2000) showing lower Mn amounts in the soil and plant parts at 0.5 km than further away from the Harjavalta smelter. The results showed that the heavy metal pollution had no significant effect on the localization of P, S, K and Ca in different cell compartments, except the amounts of Ca were slightly impared near the smelter, while K amounts were elevated. Exchangeable Ca and K concentrations in the organic soil are clearly decreased near the smelter (Derome and Lindroos. 1998) but the total nutrient (Ca. K. P and S) concentrations in E. nigrum are approximately the same or higher near the smelter than in 8 km distance from the smelter. In contrast, the Mg concentrations in stems and leaves of E. nigrum (Uhlig el ah, 2000) and 15 the exchangeable Mg concentrations in the organic soil (Derome and Lindroos. 1998) are clearly depressed near the smelter and e.g. pines are suffering from Mg deficiency (Derome and Nieminen. 1998). Mg amounts obtained with EDXS in the stems were near to or below the detection limit at both sites and therefore did not indicate any Mg deficiency especially near the smelter. In the present study the highest element (Ca, P, S, Al. As) peaks were detected in the electron-dense material, but the elemental composition and the localization of this material varied considerably. Earlier, an accumulation of metals (e.g. Cu. Zn) and nutrients (K. Ca. P) in electron-dense material in vacuoles has been reported (e.g. Vazquez, et al., 1994; Neumann et al., 1995). According to light microscopical investigations, these vacuolar precipitates in ray cells and phloem mainly consisted of phenolic material. Carlquist (1989), who investigated the wood anatomy of E. nigrum. observed that electron-dense material occurs most abundantly in rays, but also in tracheids and vessels. In this study, there was more frequently electron-dense material in the stem samples originating from 0.5 km than from 8 km from the smelter and the material was distributed throughout the stem tissue. Phenolic substances have a specific function in the ecology of E. nigrum. The exudation of phenolics by E. nigrum leaf trichomes (Wollenweber et al., 1992) inhibits the establishment of tree seedlings (e.g. Nilsson and Zackrisson, 1992; Zackrisson and Nilsson. 1992; Nilsson, 1994). Due to the higher frequence of this electron-dense material in the stems from the highly polluted site and the high heavy metal amounts in this material measured by EDXS. it can be assumed that this phenolic, electron-dense material has a special function in the heavy metal tolerance of E. nigrum. In this study. Cu was rather homogeneously distributed in the tissue and occurred in cell walls, vacuoles and cytoplasm of the living tissue as well as in the lumen of the vascular tissue. Recent electron microscopical studies of Silene vulgaris leaves showed that Cu is located in epidermal cell walls bound to glycoprotein with oxalate oxidase activity (Bringezu et al., 1999). The Cu tolerance of another dwarf shrub Calluna vulgaris (Hull.) was supposed to be achieved by an accumulation of Cu in the ericoid mvcorrhizas of the roots preventing the metal transport to the shoots (Bradley et al., 1981; 1982). In contrast to that, this study as well as previous greenhouse studies (Monni et al., 2000 a) showed that in E. nigrum also a root-to-shoot transport occurs. 16 In the xylem sap more than 98% of the Cu is present in complexed form (Graham. 1979), where Cu has been suggested to bind to amino acids (White et al., 1981). Copper has a high affinity to cysteine-rich proteins, carboxylic and phenolic groups (Marschner. 1995) and the connection of elevated Cu amount and proteins and vacuolar phenolic compounds in leaves and roots has been found (e.g. Rauser. 1984; Neumann et al., 1995; Ernst et al., 2000). The S amounts in this study, however, did not indicate any connection of Cu to proteins. EDXS does not detect N, which might have provided more information about the protein or phenolic amounts in cell walls, vacuoles or cytoplasm. It was not shown whether phytochelatins. as having high affinity for Cu (e.g. Marschner. 1995). or other complexing agents play a role in heavy metal tolerance of E. nigrum. However, because of the homo geneous distribution of Cu in the tissue, it seems that not only one specific mechanism is involved in the Cu tolerance of E. nigrum. The high affinity of Fe for ligands like organic acids and inorganic phosphate makes it unlikely that ionic forms of Fe has any importance for the short- and long-distance transport in plants (Marschner, 1995). Like for Cu, also Fe accumulation in ericoid mycorrhizal roots of C. vulgaris has been suggested to be responsible for the reduction of Fe transport to the shoots when exposed to high Fe concentrations (Shaw et al., 1990). The results of this study, however, showed that also Fe was translocated to the shoot. Fe has been reported to be bound to epidermal cell walls of tolerant Silene vulgaris leaves (Bringezu et al., 1999). whereas cytoplasm, vacuoles and cellular organelles of tolerant Minuartia verna and Silene vulgaris leaves contained only traces or no Fe (Neumann et al., 1997; Bringezu, et al., 1999). In this study, the Fe amount in cell walls especially of ray cells and sclereids was higher near the smelter. Also the localization in the cytoplasm and vacuoles indicated a detoxification of Fe in these cell compartments. The specific complexing agents could not be shown according to this study. Unlike other elements As and A 1 are not essential for plants. In the highly contaminated samples the primary cell walls of ray cells and phloem had higher As amounts than that of dead cells (xylem, sclereids) indicating a placement of As to the apoplast of living cells. An accumulation of metals in vacuoles of the contaminated stems was also observed. The cellular localization of As has been researched in the wood of Pin us svlvestris after 17 chromated copper arsenate (CCA) treatment. All the cell compartments were not studied but the results showed that lumen and middle lamella of tracheids and ray cell parenchyma of the peeling samples and sapwood contained As, whereas in untreated samples no As could be found (Helsen and Van den Bulck. 1998). In tolerant Silene vulgaris leaves A 1 has been bound to epidermal cell walls (Bringezu et al, 1999). whereas cytoplasm, vacuoles and cellular organelles of tolerant Minuartia verna and Silene vulgaris leaves contained no Al (Neumann et al., 1997; Bringezu, et al., 1999). In this study the effect of pollution was not clearly seen and Al was detected both in apoplast and symplasm. the significant differences between the sites being only seen in cells walls or cytoplasm in few tissues as well as in the electron dense material. However, the differences were very small and the overall amount was almost the same in all cell compartments. Therefore the similar pattern as above was not clearly seen according to this study. In plants. Mn is mainly in ionic form as it forms unstable complexes with organic ligands (Marschner, 1995). As being the cofactor for different enzymes (Burnell, 1988) Mn was abundant in the cytoplasm of control plants. However, in samples near to the smelter such a pattern could not be found. Because of the extremely low amounts of Mn near the smelter, some Mn deficiency might occur. Nutritional imbalance might be the reason for the decreased metabolic efficiency, and thus the decreased contents of organic acids (malic and citric acids) and chlorophyll in E. nigrum near the smelter. However, also decreased Mg contents as well as increased Fe contents near the smelter might affect the chlorophyll pigment contents of E. nigrum (Monni et al., 2000b). In this study. Ca was mainly located in cell walls and cytoplasm of living cells (ray cells, phloem), but also in the torus of pits between tracheids and vessels indicating its significance as an important constituent of the plant middle lamella and cell wall. Ca is mainly bound to structural material, and should be most abundant in cell walls (Kirkbv and Pilbeam. 1984). while the amount of Ca in the cytoplasm is usually very low (Marschner. 1995). However, e.g. environmental stresses or ABA can induce an increase in cytosolic free Ca 2+ (e.g. Marschner, 1995). However, this was not found near the smelter where ABA in current-year leaves and stems is increased (Monni et al., 2000b) and cytosolic Ca decreased. 18 K is very mobile and mainly occurs in the symplasm. the cytosolic K content being usually relatively constant (Hsiao and Läuchli, 1986; Marschner, 1995). In E. nigrum, K occurred mainly in the cytoplasm, but the amounts varied due to the site or tissue. Also in primary walls of the phloem and primary ray cells high K contents were found. K amounts were low and relatively constant in vacuoles and lumens of E. nigrum stems independent on tissue or site. According to Hsiao and Läuchli (1986). the K amounts of vacuoles may vary whereas in the apoplast K amounts are usually very low (Läuchli and Pfluger. 1978; Hsiao and Läuchli. 1986; Marschner. 1995). The dead cells (tracheids, vassels, sclereids) had only low amounts of P, the P being concentrated in the electron-dense material of the ray cells and phloem and cytoplasm, where it regulates metabolic pathways. It is also a constituent of nucleic acids and ATP. S. being an important component of amino acids, proteins and (Marschner. 1995) heavy metal binding phytochelatins (Marschner. 1995; Keltjens and van Beusichem. 1998). was also located in the cytoplasm in the high and low contaminated samples. Heavy metal pollution did not cause any change in the pattern of the S and P amounts in different cell organelles. 5. Conclusions There were higher amounts of Cu, As and Fe in the cell compartments of E. nigrum at 0.5 km than at 8 km. The A 1 and Zn amounts, in contrast, did not differ significantly between the two sites. Cu was localized in several cell compartments (cell walls, cytoplasm, vacuoles, lumens) whereas the As amounts were higher in the primary cell walls of living (ray cells, phloem) than of dead cells (xylem, sclereids). Elevated As amounts were also found in cytoplasm and vacuoles. Ray cells, phloem and sclereids had elevated amounts of Fe compared to the other tissues in the contaminated stem samples but. owing to the high variation between the replicates, generally no significant differences were found. The vacuoles, cytoplasm and cell walls of the contaminated stems also had higher Fe amounts indicating the transport of metals to these cell compartments. Based on the rather homogeneous localization of Cu. As and Fe in the living tissue and increased amounts of Cu. As and Fe in vacuoles, cell walls and cytoplasm near the smelter, it seems that not only one specific mechanism contribute to the heavy metal tolerance of E. nigrum. It was not shown whether phytochelatins and organic acids are involved in heavy metal tolerance in 19 the cytoplasm and vacuoles, respectively, as shown in earlier investigations. According to light microscopical examinations, electron-dense material consisted partly of phenolic material, and it occurred more frequently in the polluted samples. Therefore the electron dense material containing phenolic substances might have a function in the heavy metal tolerance of E. nigrum. Acknowledgements We wish to thank the staff at the Department of Electron Microscopy. University of Helsinki, for the practical work and Jyrki Juhanoja for valuable advice during the study. We also thank the staff at the University of Bremen and University of Tubingen for the practical help. John Derome from the Finnish Forest Research Institute revised the English, which is greatly acknowledged. The work was supported by the Maj and Tor Nessling Foundation and the Alfred Kordelin Foundation. 20 References Antonovics. J., Bradshaw. A.D. and Turner, R.G., 1971. Heavy metal tolerance in plants Adv. Ecol. Res., 7, 1-85. Baker. A.. 1981. Accumulators and excluders - strategies in the response of plants to heavy metals. J. Plant Nutr., 3, 643-654. Baker. A. and Walker. P., 1990. Ecophysiology of metal uptake by tolerant plants. In: Heavy metal tolerance in plants: Evolutionary aspects. Shaw, J. (ed.). CRC Press. FL. pp. 155-178. Bell. J. and Tallis. J., 1973. Biological flora of the British Isles. Empetrum nigrum. J. Ecol., 61,289-305. Bradley, R.. Burt, A.J.. Read. D.J., 1981. Mycorrhizal infection and resistance to heavy metal toxicity in Calluna vulgaris. Nature, 292, 335-337. Bradley, R., Burt. A.J.. Read. D.J.. 1982. The biology of mycorrhiza in the Ericaceae. VIII. The role of mycorrhizal infection in heavy metal resistance. New Phytol.. 91, 197- 209. Bringezu, K.. Lichtenberger. 0., Leopold. I. and Neumann. D., 1999. Heavy metal tolerance of Silene vulgaris. J. Plant Physiol., 154. 536-546. Burnell, J.N., 1988. The biochemistry of manganese in plants. In: Manganese in soils and plants, Graham. R.D.. Hannam, J., Uren, N.C. (Eds), Kluwer Academic Publishers. Dordrecht. The Netherlands, pp. 125-133. Bucking, H., Beckmann. S., Heyser, W. and Kottke. 1.. 1998. Elemental contents in vacuolar granules of ectomvcorrhizal fungi measured by EELS and EDXS. A comparison of different methods and preparation techniques. Micron, 29, 53-61. Carlquist. S.. 1989. Wood and bark anatomy of Empetraceae\ comments on paedomorphosis in woods of certain small shrubs. Aliso, 12, 497-515. Chertov, 0.. Nadporozshskaya. M.. Lapshina. I. and Grigorieva. 0.. 1993. Soil degradation in surroundings of "pechenganikel" smelter complex. In: Aerial pollution in Kola Peninsula, Kozlov. M. V.. Haukioja. E. and Yarmishko, VT. (Eds). Proceedings of the international workshop, 14-16 April, St Petersburg, pp. 159-162. Derome. J., 2000. Effects of heavy-metal and sulphur deposition on the chemical properties of forest soil in the vicinity of a Cu-Ni smelter, and means of reducing the detrimental effects of heavy metals. Doctoral thesis, Finnish Forest Research Institute. Research Papers 769, Finland. Derome. J. and Lindroos. A.-J., 1998. Effects of heavy metal contamination on macronut rient availability and acidification parameters in forest soil in the vicinity of the Harjavalta Cu-Ni smelter, SW Finland. Environ. Poll., 99, 225-232. Derome, J., Nieminen. T., 1998. Metal and macronutrient fluxes in heavy-metal polluted Scots pine ecosystems in SW Finland. Environ. Poll., 103, 219-228. Ernst. W.H.0.. Nelissen. H.J.M. and Ten Bookum. W.M., 2000. Combination toxicology of metal-enriched soils: physiological responses of a Zn- and Cd-resistant ecotvpe of Silene vulgaris on polymetallic soils. En v. and Exp. Botany. 43, 55-71. Fritz. E.. 1980. Microradioautographic localization of assimilates in phloem: problems and new method. Ber. Dtsch. Bot. Ges.. 93, 109-121. Graham. R.D.. 1979. Transport of copper and manganese to the xylem exudate of sunflower. Plant Cell Environ. 2, 139-143. Helmisaari, H.-S., Derome. J.. Fritze, H., Nieminen, T., Palmgren, K.. Salemaa, M. and Vanha- Majamaa. 1.. 1995. Copper in scots pine forests around a heavy-metal smelter in South- Western Finland. Water, Air and Soil Poll., 85, 1727-1732. 21 Hclsen. L. and Van den Buick. E.. 1998. The microdistribution of copper, chromium and arsenic in CCA treated wood and its pvrolysis residue using energy dispersive x ray analysis in scanning electron microscopy. Hohforschung, 52, 607-614. Hsiao. T.C.. and Läuchli, A., 1986. Role of potassium in plant-water relations. In: Advances in Plant Nutrition. Tinkler 8.. Läuchli. A. (Eds). Vol 2. pp. 281-312 Praeger Scientific. New York. Keltjens. W.G. and van Beusichem, M.L.. 1998. Phvtochelatins as biomarkers for heavy metal stress in maize (Zea mays L.) and wheat (Triticum aestivum L.): combined effects of copper and cadmium. Plant Soil, 203, 119-126. Kirkby. E. A.. Pilbeam. D.J.. 1984. Calcium as a plant nutrient. Plant Cell Environ., 7, 397-405. Kottke. 1.. 1994. Localization and identification of elements in mycorrhizas. Advantages and limits of electron energy loss spectroscopy. Acta Bot. Gallica, 141, 507-510 Laaksovirta, K. and Silvola. J., 1975. Effect of air pollution by copper, sulphuric acid and fertilizer factories on plants at Haijavalta, W. Finland. Ann. Bot. Fenn., 12, 81-88 Lee. J.. Reeves. R.D.. Brooks, R.R. and Jaffre T., 1978. The relation between nickel and citric acid in some nickel-accumulating plants. Phvtochemistry, 17, 1033-1035. L'Huillier. L., d'Auzac. J.. Durand. M. and Michaud-Ferriere, N., 1996. Nickel effects on two maize (Zea mays) cultivars: growth, structure. Ni concentration, and localization. Can. J. Bot.. 74. 1547-1554. Lichtenberger, O. and Neumann. D., 1997. Analytical electron microscopy as a powerful tool in plant cell biology: examples using electron energy loss spectroscopy and x-ray microanalysis. Eur. J. Cell Biol., 73, 378-386. Läuchli. A., and Pfliiger. R., 1978. Potassium transport through plant cell membranes and metabolic role of potassium in plants. Proc. 11th Congr. Int. Potash Inst. Bern, pp.lll-163. Marschner. H., 1995. Mineral nutrition of higher plants. 2nd edn. London, UK: Academic press Ltd. pp. 333-347. Miller. H., 1975. Anatomical characteristics of some woody plants of the Angmagssalik district of southeast Greenland. Meddelelser om Gronland, 198, 1-30. Monni. S.. Salemaa. M. and Millar, N.. 2000 a. The tolerance of Empetrum nigrum to copper and nickel. Environ. Poll., 109, 221-229. Monni. S.. Uhlig, C., Hansen. E.. Magel. E.. 2000b. Ecophysiological responses of Empetrum nigrum to heavy metal pollution. Environ. Poll., 112, (in press) Mullins, M.. Hardwick. K. and Thurman. D. 1985. Heavy metal location by analytical electron microscopy in conventionally fixed and freeze-substituted roots of metal tolerant and non tolerant ecotvpes. In: Proceedings of international conference on heavy metals in the environment. Athens, pp. 43-46. Neumann. D.. Nieden, U.. Lichtenberger, O. and Leopold. 1., 1995. How docs Armeria maritima tolerate high heavy metal concentrations. J. Plant Physiol., 146, 704-717 Neumann. D.. Nieden. U.. Schwieger. W.. Leopold. I. and Lichtenberger. 0.. 1997. Heavy metal tolerance ofMinuartia verna.,/. Plant Physiol., 151, 101-108. Nilsson. M.-C., 1994. Separation of allelopathy and resource competition by the boreal dwarf shrub Empetrum hermaphroditum Hagerup. Oecologia, 98, 1-7. Nilsson, M.-C. and Zackrisson, 0.. 1992. Inhibition of Scots pine seedling establishment by Empetrum hermaphroditum. J. Chem. Ecol., 18, 1857-1870. Proctor. J.. 1971.The plant ecology of serpentine. 11. Plant response to serpentine soils. J. Ecol.. 59, 397-410. 22 Proctor. J. and Woodell. S.R.J.. 1971. The plant ecology of serpentine. I. Serpentine vegetation of England and Scotland. J. Ecoi. 59, 375-395. Rantalahti. R.. 1995. Outokummun Harjavallan tehtaiden ympäristönsuojelusta Vaskeriviesti, 3+4. 2-5. Rauser. W.E.. 1984. Copper-binding protein and copper tolerance in Agrostis gigantea Plant Science Letters, 33, 239-247. Robinson. D.G.. Ehlers, U.. Herken. R.. Herrmann. 8.. Mayer. F. and Schurmann, F.-W.. 1985. Präparationsmethodik in der Elektronenmikroskopie. Springer. Berlin. Shaw. G.. Leake, J.R., Baker. A.J.M., Read, D.J., 1990. The biology of mvcorrhiza in the Ericaceae XVII. The role of mycorrhizal infection in the regulation of iron uptake by ericaceous plants. New Phytol., 115, 251-258. Sokal. R.R. andßohlf, F.J.. 1995. Biometry. W. H. Freeman. New York Spurr. A.. 1969. A low-viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res., 26, 31-43. Statistix. 1996. Statistix for windows. User's manual. Borland International Inc.. Tallahas see, FL. Stelzer. R. and Lehmann. H.. 1993. Recent developments in electron microscopical techniques for studying ion localization in plant cells. Plant Soil. 155/156, 33-43 Turnau. K, Kottke. I. and Oberwinkler. F.. 1993 a. Element localization in mycorrhizal roots of Pteridiwn aquilinum (L.) Kuhn collected from experimental plots treated with cadmium dust. New Phytol.. 123, 313-324. Turnau. K., Kottke, I. and Oberwinkler, F., 1993b. Paxillus involutus-Pinus sylvestris mycorrhizae from heavily polluted forest. I. Element Localization Using Electron Energy Loss Spectroscopy and Imaging. Bot. Acta, 106, 213-219. Uhlig. C.. Salemaa. M., Vanha-Majamaa. 1.. Derome, J.. 2000. Element distribution in Empetrum nigrum microsites at heavv-metal contaminated sites in Harjavalta. Western Finland. Environ. Poll, (in press). Vazquez. M.D.. Poschenrieder, Ch.. Barcelö, J., Baker. A.J.M., Hatton. P and Cope. G.H.. 1994. Compartmentation of zinc in roots and leaves of the zinc hyperaccumulator Thlaspi caerulescens J & C Presl. Bot. Acta, 107, 243-250. Verklcij. J.A.C. and Schat. H., 1990. Mechanisms of metal tolerance in higher plants. In: Heavy metal tolerance in plants: Evolutionary aspects. Shaw. J. (ed.). CRC Press. FL. pp. 179-193. White. M.C., Decker. A.M.. Chaney, R.L.. 1981. Metal complexation in xylem fluid. I. Chemical composition of tomato and soybean stem exudate. Plant Physiol., 67, 292-300. Wollenweber, E.. Dörr. M., Stelzer. R. and Arriaga-Giner, F.J.. 1992. Lipophilic phenolics from the leaves of Empetrum nigrum - Chemical structures and exudate localization. Bot. Acta, 105, 300-305. Wollgiehn. R. and Neumann, D., 1999. Metal stress response and tolerance of cultured cells from Silene vulgaris and Lycopersicon peruvianum: Role of heat stress proteins. J. Plant Physiol., 154, 547-553. Woolhouse. H. W., 1983. Toxicity and tolerance in the responses of plants to metals. In Encycl. Plant Physiol., pp. 245-300. Yang. X., Baligar, V.. Foster, J. and Martens, D.. 1997. Accumulation and transport of nickel in relation to organic acids in ryegrass and maize grown with different nickel levels. Plant Soil, 196, 271-276. 23 Zackrisson. O. and Nilsson. M.-C., 1992. Allelopathic effects bv Empetrum hermaphrodi tum on seed germination of two boreal tree species. Can. J. For. Res., 22, 1310- 1319. ISBN 951-40-1749-8 ISSN 0358-4283 Hakapaino 2000