Aspects of the ecology of mat-forming lichens

Lichen species in the genera Cladonia (subgenus Cladina), Cetraria, Stereocaulon and Alectoria are important vegetation components on well-drained terrain and on elevated micro-sites in peatlands in boreal-Arctic regions. These lichens often form closed mats, the component thalli in which grow vertically upwards at rhe apices and die off in the older basal regions; they are therefore only loosely attached to the underlying soil. This growth habit is relatively unusual in lichens being found in <0.5% of known species. It might facilitate internal nutrienr recycling and higher growth rates and, together with the production of allelochemicals, it might underlie the considerable ecological success of mar-forming lichens; experiments to critically assess the importance of these processes are required. Mat-forming lichens can constitute in excess of 60% of the winter food intake of caribou and reindeer. Accordingly there is a pressing need for data on lichen growth rates, measured as mass incremenr, in order to help determine the carrying capacity of winter ranges for rhese herbivores and to better predict recovery rates following grazing. Trampling during the snowfree season fragments lichen thalli; mat-forming lichens regenerate very successfully from thallus fragments provided trampling does nor re-occur. Frequent recurrence of trampling creates disturbed habitats from which lichens wil l rapidly become eliminated consistent with J.P. Grime's CSR strategy theory. Such damage to lichen ground cover has occurred where reindeer or caribou are unable to migrate away from their winter range such as on small islands or where political boundaries have been fenced; it can also occur on summer range that contains a significant lichen component and on winter range where numbers of migrarory animals become excessive. Species of Stereocaulon, and other genera that contain cyanobacteria (most notably Peltigera and Nephroma), are among the principal agents of nitrogen fixation in boreal-arcric regions. Stereocaulon-dominated subarctic woodlands provide excellent model systems in which to investigate the role of lichens in nitrogen cycling. Mat-forming lichens are sensitive indicators of atmospheric deposition partly because they occur in open situations in which they intercept precipitation and particulates directly with minimal modification by vascular plant overstoreys. Data from both the U K and northern Russia are presented to illustrate geographical relationships between lichen chemistry and atmospheric deposition of nitrogen and acidity. The ecology of mat-fotming lichens remains under-researched and good opportunities exist for making significant contributions to this field including areas that relate directly to the management of arctic ungulates. K e y words : acid rain, allelopathy, grazing, growth, lichens, nitrogen fixation, recycling, trampling. Rangifer, 20 (2-3): 127-139


Introduction
That lichens are eaten by reindeer and caribou is one of the better known facts about lichens amongst lay people of mid latitudes.It is also quite widely understood that lichens are symbiotic organisms and not plants in the strict sense.Although they are primary producers, and frequently have dimensions of small vascular plants, they are in fact composed of fungi deriving nutrition (carbon and sometimes nitrogen) from unicellular or filamentous algae and/or cyanobacteria in mutualistic symbiotic associations.Accordingly, it is strictly inappropriate to refer to lichens as organisms but for the sake of brevity it is often convenient to do so.Lichen-forming fungi constitute about one fifth of the fungal kingdom (Hawksworth et al., 1995) and there is evidence that lichens represent an ancient association that was present amongst terrestrial crust communities during the early to mid Paleozoic Era (Shear, 1991;Taylor et al., 1995).Yet because of their minor economic importance the biology of lichens remains poorly understood in comparison to the other principal types of terrestrial mutualistic "plant" symbioses (i.e.mycorrhizas and legumes/ actinorrhizas).Lichens grow on substrata that are deficient in nitrogen and phosphorus, and that are subject to frequent episodes of acute drought.Such substrata are relatively abundant in boreal-arctic regions.In particular, the poorly developed to skeletal soils typical of high latitudes frequently support lichenrich ground cover.According to Larson (1987) lichen-dominated vegetation covers approximately 8% of the world's terrestrial surface and much of this is in the Subarctic and Arctic.However, factors that control the development of lichen-dominated terrain remain incompletely understood.Such vegetation appears to be largely restricted to welldrained surfaces, including elevated microsites in peat bogs (Kershaw, 1977;Foster & Glaser, 1986;Ahti & Oksanen, 1990); these are sites in which vascular plant vigour is reduced due to low soil fertility and/or that are at an early successional stage following fire.Precipitation does not appear to 128 determine the frequency of lichen-dominated terrain; for example, lichen woodland is abundant in both north-central Canada (c. 300 mm annual precipitation) and Labrador (c.800 mm) although the relative importance of particular lichen species may differ (Kershaw, 1977).The most ecologically successful terricolous lichens are mat-forming species of the genera Cladonia (subgenus Cladina), Cetraria, Stereocaulon, and Alectoria (Fig. 1).Certain key species in these genera can form extensive ground cover in subarctic lichen woodland, forest tundra and tundra heathlands.They form the principal component of lichen intake by reindeer and caribou on their winter ranges (e.g.Miller, 1976;Holleman & Luick, 1977;Boertje, 1990), although epiphytic species are also important in the diet of some populations (e.g.Rominger et al., 1994).This review focuses on aspects of the ecology of mat-forming lichens.

The mat-forming habit
Mat-forming lichens grow acropetally (i.e. at their apices) and vertically upwards while in mature mats Rangifer, 20 (2-3), 2000 the thallus bases die off.Thus the upper living parts of the mats are often supported physically by a deep layer of dead, structurally intact, thallus or necromass.This growth habit is relatively unusual in lichens being found in <0.5% of the 13 500 known species; the vast majority of lichens are firmly attached to their substrata.Precipitation and dry deposition of dust particles and gases are believed to be the primary sources of N and P for lichen growth.Thus in deep mats, zones of depletion develop in the lower strata due to interception of light and nutrients by the surface layers.Several authors have discussed the horizontal stratification of microclimate and thallus morphology that occurs within lichen mats (Carstairs & Oechel, 1978;Lechowicz, 1983;Pakarinen, 1985;Sveinbjornsson, 1987;Coxson & Lancaster, 1989;Crittenden, 1991).Thallus below the light compensation depth (i.e. the depth at which photosynthetic C0 2 assimilation equals respiratory loss of C0 2 ) represents a respiratory burden and is therefore expendable.In the northern hemisphere, Cladonia stellaris is the supreme example of a mat-forming lichen.
Lichens, being essentially fungal mycelia, are poikilohydric, i.e. their water status is coupled to that of the atmosphere.Under the influence of evaporative forces lichens lose water becoming dry and largely metabolically inactive.A widely held view is that lichens are C-limited because photosynthetic activity is restricted to intermittent periods of adequate water availability.However, there is evidence that the more productive mat-forming lichens are N and/or P limited (Crittenden et al., 1994).Several of the above authors have suggested that N and P in senescing basal parts of lichen thalli might be remobilized and translocated to the growing apices.Such internal recycling would be ecologically advantageous because it would permit a higher growth rate at the apices than could be sustained by new atmospheric supplies alone.This, in turn, would result in a higher rate of necromass production and deeper mats with the capacity to cast deep shade.In vascular plant ecology the development of a tall closed canopy is a mechanism by which some plant species exclude potential competitors and exert dominance (Grime, 1979).The capacity to recycle nutrients, which would be dependent on the loss of attachment between living thallus and the substratum, might underlie the ecological success of mat-fofming lichens.
There is conflicting information on the occurrence of viable seed banks beneath lichen ground RangiSer, 20 (2-3), 2000 cover.Johnson (1975) failed to find germinable seeds in spruce-lichen woodland soils in the Northwest Territories in Canada whereas I have found abundant viable seeds beneath mats of Cladonia stellaris and Stereocaulon paschale in Finn¬ marksvidda in Norway.It is probable that seeds that descend to the bottom of mats will suffer lightdeprivation sufficient to inhibit either germination or seedling growth (Kershaw, 1977) while Allen (1929) described seedlings in lichen mats being uprooted or damaged by the expansion of dry mats upon rehydration.Thus vascular plant species reproducing by small seeds may be excluded from well-developed lichen mats.However, studies in Sweden (Steijlen et al., 1995;Zackrisson et al., 1995) have shown that Cladonia-dommaxed ground cover in coniferous forests is a more favourable environment for the establishment and survival of spruce and pine seedlings than vegetation dominated by either Etnpetrum hermaphroditum or Pleurozium schreberi.Germination of pine and spruce seeds is not completely suppressed by light deprivation (Sarvas, 1950) and it is possible that a larger seed size might sustain the growth of pine and spruce seedlings through the lichen canopy into the 'photic' zone.However, Cowles (1982) found the centres of deep lichen cushions to be formidable barriers to the establishment of spruce in lichen woodland in northern Quebec whereas, in contrast, polygonal cracks that form in desiccated mats were favourable sites for establishment.

Allelopathic activity
Lichens produce a diverse range of secondary chemical products, many of them phenolic derivatives, and most being unique to lichens (Fahselt, 1994).Such products probably have several functions in the lichen thallus.Since many have antimicrobial properties, and/or are effective in deterring potential herbivores, it seems reasonable to suggest that one of their functions is chemical defence of the thallus (Lawrey, 1986(Lawrey, , 1989)).The yellow compound usnic acid is one of the more potent lichen products in terms of antimicrobial activity and is produced commercially as a pharmaceutical (Vartia, 1973); it occurs in a large number of lichen genera and is a typical compound for the subgenus Cladina; for example, it contributes about 2% of the dry mass of C. stellaris (Huovinen & Ahti, 1986).Although most lichen secondary products are only sparingly soluble in water, it has been suggested that trace quantities released into the soil from healthy thalli during rainfall, and from decomposing thalli, might inhibit the growth of neighboring vascular plants.Species of Cladonia (subgenus Cladina) have been reported to retard growth and mycorrhizal development in boreal forest trees (Brown & Mikola, 1974;Fisher, 1979).This is an attractive idea because such allelopathic effects provide yet another mechanism by which a plant can reduce the growth rate of its neighbours and exert dominance.However, despite a sizeable literature on this topic (reviewed by Lawrey, 1984;1986) the importance of allelopathy in the ecological success of mat-forming lichens still remains to be critically assessed.
There are several potential pitfalls in experiments designed to test for allelopathic interactions between plants (Harper, 1975;Qasem & Hill, 1989).One of these is that quantities of metabolites to which test plants are exposed can greatly exceed those occurring in nature so that reported toxic effects can be exaggerated, if not entirely a product of the experimental conditions.Thus lichen extracts prepared using organic solvents (e.g.Goldner et al., 1986) or by homogenizing thalli in water (e.g. Brown & Mikola, 1974;Kaitera et al, 1996) may contain ecologically improbable concentrations of metabolites.Cowles (1982) conducted a field experiment in northern Quebec, that has been frequently cited in the literature, to investigate the effects of lichens on the growth of spruce.He removed lichen ground cover from plots in lichen woodland and found that this had a negative effect on the growth of black spruce as indicated by a reduction in the rate of branch elongation.This negative effect might have resulted from enhanced évapotranspiration since it could be counteracted by covering lichen-free plots with polyethylene sheet.Cowles also investigated the potential role of lichen allelopathy by soaking lichen mats in barrels of water for 24 h and then applying the filtrate to lichen-free plots covered with polyethylene sheet to compensate for the lost insulating effect of the lichens.This procedure was repeated daily during the snow-free seasons for 4 y and resulted in a negative growth response of broadly similar magnitude to the positive effect of lichen cover.However, the author does not indicate the frequency with which lichen in the barrels was renewed during the experiment; water that has passed through a dense mass of dead, and possibly fermenting, lichen is likely to differ markedly in chemical composition from rain-fall that has percolated through a living lichen stand.It should also be remembered that lichens are not the only producers of allelochemicals in these ecosystems: evidence suggests that ericaceous shrubs (Pellissier, 1993;Nilsson et al., 1993;Inderjit & Mallik, 1997;Yamasaki et al, 1998) and feather mosses (Steijlen et al, 1995) also produce phytotoxins that inhibit the establishment and/or growth of trees (see discussion of seed banks above).P. D. Crittenden and R. P. Beckett (unpubl.)investigated the effects of lichen-modified rainwater on mycorrhizal development and growth in the cultivated cranberry (Vaccinium macrocarpon) grown in sand-solution culture.Mats of Cladonia stellaris and Stereocaulon paschale were reconstructed on stainless steel grilles superimposed over rain gauges in lichen-rich birch woodland at Kevo, northern Finland, following the method of Crittenden (1983).Water that percolated through the mats during natural rainfall events was collected quantitatively, filtered to pass 0.22 pm and applied in 5 ml aliquots to the surface of the sand surrounding each Vaccinium plant; these additions were repeated at intervals of several hours until all water that had been collected during a rainfall event was exhausted.Control plants received either unmodified rainfall or deionized water, and the Vaccinium was grown either in the mycorrhizal (inoculated with the fungus Hymenoscyphus ericaea [Smith & Read, 1997]) or non-mycorrhizal states and under either low or high nitrogen regimes (2.7 or 27 parts N/10 6 , respectively, in applied nutrient solutions).The only statistically significant effect of lichen-modified rainwater during a three month growth period was an increase in the extent of mycorrhizal infection in the low nitrogen treatment (Fig. 2).In this experiment the quantities of lichen products added to the experimental plants were probably ecologically relevant.The two lichen species supported over the replicate rain guages appeared to remain in a healthy state: they increased in dry mass during the experiment and did not release K* into the percolating rain water, indicating cell membrane integrity (see Buck & Brown, 1979;Tarhanen et al, 1996).The promorion of ericoid mycorrhizal development may have been due to utilization by the mycorrhizal fungus of exogenous organic carbon present in the lichen-modified rainfall thus promoting more effective colonization of the rhizosphere prior to establishing infection points.It is debatable whether this effect would be evident in podzolic soils containing abundant soluble organic compounds.Nonetheless, Fig. 2

Co-existence with vascular plants
Terricolous lichens grow on oligotrophic sandy or podzolic soils on at least moderately well-drained terrain.Vascular plants adapted to these conditions have low growth rates (Grime, 1979) and have the capacity to absorb simple and complex organic forms of nitrogen either directly (Kielland, 1994; Schimel & Chapin, 1996) or via mycorrhizal associations (Smith & Read, 1997;Nasholm et al, 1998), thus largely by-passing the need for microbiologically-driven mineralization processes.Vascular plants show plasticity in their root/shoot ratio and typically develop larger root systems in nutrient deficient soils (e.g.Chapin, 1980;Tilman & Wedin,1991) increasing the carbon cost of mineral nutrient recovery.Clearly, with increasing nutrient scarcity there is a limit to how "rooty" a plant can become given the increased respiratory cost of Rangifer, 20 (2-3), 2000 expanding below-ground biomass.As this limit is approached and plant vigour declines, a point will be reached at which productive terricolous matforming lichens can co-exist with vascular plants (cf.Manseau et al, 1996).
Lichens do not have root systems and are probably almost entirely dependent on atmospheric deposition for supplies of N and P (Crittenden et al, 1994).The seasonal intake of these elements by lichens may be substantially less than their vascular plant neighbours but the carbon cost of this acquisition is probably small in comparison.Thus the growth rates of above-ground phytomass in lichens and vascular plants could be similar or, in extremely nutrient poor soils, greater in lichens.Being rootless, lichens are largely uncoupled from soil water and thus must tolerate periods of desiccation during which metabolic activity is minimal.Lichen ecophysiologists tend to view desiccation negatively since it is believed that this limits lichen growth rates.In addition, there is evidence that rewetting of lichen thalli (e.g. during the onset of rainfall) is itself associated with a carbon cost (e.g.Lechowicz, 1981;Dudley & Lechowicz, 1987).However, because precipitation is one of the major processes that delivers nutrients to lichens, periods of desiccation between infrequent precipitation events might decrease the carbon cost of nutrient capture.However, fertilization of these oligotrophic communities can increase plant growth rates and result in the competitive exclusion of lichens (e.g.Persson, 1981;Jonasson, 1992).It is quite common in the Subarctic for well-drained terrain to experience acute summer droughts that could be lethal to fine roots of trees and dwarf shrubs which develop close to the soil surface.Thus, in addition to paucity of soil mineral nutrients, periodic drought may also be a significant factor suppressing vascular plant vigour and promoting the development of lichen ground cover.

Lichen growth and the impact of trampling and grazing
Terricolous lichens typically constitute in excess of 60% of the winter food intake by reindeer and caribou (Miller, 1976;Boertje, 1990).Lichen consumption declines markedly during the summer months when graminoids and dwarf shrubs comprise the major food items.Crête et al. (1990) estimated that the average lichen biomass on the winter range of the Rivière George caribou herd (c.680 000 individuals in 1988) in northern Quebec averaged 1223 kg ha -1 and that in 1987 0.5-0.9% of this was consumed annually.The significance of this consumption is dependent on lichen productivity; unfortunately, as the authors point out, appropriate data on lichen growth are largely lacking.Sveinbjôrnsson (1990) has discussed some of the approaches available for measuring lichen growth rate: these range from morphological analyses (that to my knowledge have never been critically evaluated) to simulation modelling based on net C0 2 -exchange measurements.Probably the best method, and possibly the only method that should be considered for longterm productivity determination, is the measurement of mass increment (Jonsdottir et al., 1998).Few investigators have measured mass-increment in lichens and only Karenlampi (1971), Hooker (1980), Crittenden (1983Crittenden ( , 1989;;summarized in Crittenden et al, 1994) and Hyvarinen & Crittenden (1998a) have done so for mat-forming lichens.The method suffers from the criticism that lichen thalli must be detached from the ground (or underlying necromass) in order to make the initial (time zero) weighing.The effect of such detachment has not been fully investigated.

132
In northern Finland, thalli of Cladonia stellaris and Stereocaulon pascbale that were pre-weighed and inserted into reconstructed lichen mats increased in dry mass by at least 26% per year (Crittenden et al, 1994).This equates with a mean relative growth rate (RGR) of c. 0.004 mg mg" 1 d 1 , a value that compares favourably with those recorded by Karenlampi (1971) for the same species using a similar method.However, Karenlampi showed that in wet years RGR for C. stellaris could rise to 0.005¬ 0.01 d' 1 .It is instructive to compare these lichen RGR values with those for vascular plants.Grime & Hunt (1975) measured the RGR of plant species in the local flora of the Sheffield region in the UK under optimum conditions of temperature and supplies of light, water and nutrients.Of the 132 species screened Urtica dioica was found to have the highest mean RGR of 0.3 d 1 and Vaccinium vitisidaea, a species that frequently co-occurs with matforming lichens in the boreal-Arctic, had the lowest value of 0.03 d 1 .When V. vitis-idaea was grown in organic top-soil collected from beneath C. stellaris mats in Finnmarksvidda, Norway, but otherwise under favourable conditions of light, temperature and water supply, mean RGR values of 0.016 d 1 were observed (Crittenden, 1989)-Thus, under harsher field conditions (e.g.exposure to periodic summer drought, lower temperatures) it is not improbable that RGR of V. vitis-idaea would approach that of C. stellaris consistent with the coexistence of the two species.The measured growth rates for C. stellaris and S. paschale are in broad general agreement with the observation that recovery of C. stellaris ground cover that is harvested by man for ornamental purposes in northern Finland takes 5-6 years (S0chting, 1984).Virtala (1992) estimated that lichen biomass in overgrazed lichen woodland in Fennoscandia could recover from 300 kg ha" 1 to 1000 kg ha 1 in about 9 years, and to 2000 kg ha 4 in 16 years.
Mat-forming lichens are relatively loosely attached to the ground.I have argued above that this loss of firm attachment to the substratum may have permitted an increased productivity in those species adopting this growth habit.Lichen species that form deep mats (e.g. C. stellaris, C. mitis, S. paschale) are generally restticted to sheltered sites experiencing comparatively low wind speeds: these are woodlands and low arctic tundra sites with at least moderate winter snow cover (Larson & Kershaw, 1975;Larsen, 1980).The extent to which the distribution of these species is linked to snow depth, and whether this link is causal, is not known.In contrast, other species such as Cetraria nivalis and Alectoria ochroleuca frequently occur on exposed sites experiencing minimal winter snow cover (Larson & Kershaw, 1975;Petzold & Mulhern, 1987;Ahti & Oksanen, 1990).Lichens beneath deep snow are afforded protection against trampling by reindeer and caribou herds on their winter ranges (Miller, 1976).In order to graze terricolous lichens the animals excavate craters in the snow.This causes mechanical damage to mats over and above the loss due to consumption; estimates of this wastage range between to 2 -10 times the lichen mass ingested (Virtala, 1992).Kauppi (1990) has shown that in middle Finland C. Jte/Z^m-dominated lichen mats that become covered with debris, such as conifer needles and bark fragments, can fully recover in 3-8 years; comparable distutbance effects might occur due to deposition of faeces and mat-fragments in areas where caribou and reindeer have cratered intensively (Miller, 1976).
In summer, dry lichen mats are destroyed rapidly by trampling (cf.Bayfield, Urquhart & Cooper, 1981).In the case of migratory herds such damage may be largely confined to migration routes (Boertje, 1990) or to summer ranges where these have a significant lichen component.For example, Manseau et al. (1996) have reported marked depletion of lichens in tundra used by migratory caribou as summer range in northern Quebec.My own observations in shrub tundra between Vorkuta and the Polar Ural Mountains in Russia suggest that reindeer herding has largely eliminated mat-forming lichens from the ground cover in this region.Here, Cladonia arbuscular can be found in most 1 m 2 quadrats randomly located on well-drained terrain but usually only in trace quantities, possibly indicating that efficient foraging is as least as important as ttampling; Moser, Nash & Thompson (1979) made a similar observation in Alaskan tundra.According to Ahti & Oksanen (1990) and Vilchek (1997), depletion of lichen cover in tundra due to reindeer herding is widespread in Russia.
Severe degradation of winter range owing to summer trampling can also occur where migratory movements are prevented by artificial barriers at political boundaries or by the stocking of small islands (Anon, 1974;Oksanen, 1978;Swanson & Barker, 1992;Vare et al, 1996).Deterioration of winter range in northern Finland, a result primarily of summer-time trampling, is particularly marked and on a scale that is evident in satellite images Rangifer, 20 (2-3), 2000 (Kayhko & Pellikka, 1994;Vare et al, 1996).Where numbers of migratory animals exceed the range carrying capacity then excessive winter-time foraging can also lead to depletion of the lichen resource.Such a situation has been described for caribou in northern Quebec (Arseneault et al, 1997) and Alaska (Moser, Nash & Thompson;1979), and for reindeeer in Finnmarksvidda (E.Gaare, pers. comm.;Kashulina et al, 1997) and Russia (Vilchek,

1997).
Mat-forming lichens appear to regenerate well following fragmentation.This is exemplified by lichen monocultures established at the Kevo Subarctic Research Station in northern Finland by P. Kallio in 1971.Kallio cleared vegetation from 1 m x 1 m plots on the floor of an open stand of Scots pine and then spread 200 g crushed thalli of a single lichen species into each.Monocultures of CI. stellaris, CI rangiferina, Ce. nivalis and S. paschale had developed by 1978 and in 1998 most species were present as luxurious mats.Many lichen-forming fungi in the genera Cladonia and Stereocaulon can be readily isolated into pure culture by plating small (60-200 um long) alga-free fragments of macerated thallus onto agar (Crittenden et al, 1995) and this may also be indicative of the importance of fragmentation as a means of dispersal and regeneration in these important lichen genera.
Kallio's lichen plots demonstrate that mat-forming lichens can regenerate following a massive disturbance if subsequently left undisturbed but where disturbance recurs frequently lichens will not recover.The reason for this is succinctly explained by Grime's (1979) CSR sttategy theory which states that no plant strategy has evolved that can facilitate survival in habitats that are subject to high levels of both stress and disturbance.Since mat-forming lichens are supremely good toleratots of environmental stress then, in accordance with Grime, they must also be intolerant of frequent disturbance.Thus lichens are rapidly eliminated from summer range by repeated trampling and will only recover if ungulate numbers are reduced (cf.Virtala, 1992).

Nitrogen fixation
Lichens that contain a cyanobacterium fix nitrogen.An important finding of the International Biological Programme's Tundra Biome investigations was that cyanobacterial lichens (principally of the genera Peltigera, Stereocaulon and Nephroma) are among the major agents of nitrogen fixation in the Subarctic and Arctic (Stewart, 1976).There has been a sufficiently large number of studies of nitro¬ genase activity in lichens to provide a relatively good understanding of which genera and species are the most active fixers (e.g.Peltigera spp.) and how rates of nitrogenase activity are controlled by environmental factors such as thallus water content, temperature and incident light (Kershaw, 1985).
But what is the ecological significance of nitrogen fixation by lichens?There have been several attempts to produce annual rates of N input by lichens in both northern boreal forests (Billington & Alexander, 1983) and tundra habitats (Gunther, 1989;Getsen et al, 1997).However, almost invariably these estimates have been derived from the results of the acetylene reduction assay, an analogue enzymatic teaction which, at best, only gives an approximate measure of absolute nitrogen fixation rates.Where feasible, annual rates of nitrogen fixation can probably be more appropriately obtained from the product of growth (as mass increment -see above) and thallus nitrogen concentration but as yet such calculations have not been made [with the exception of Crittenden (1989)}.Further, estimates of N input by lichens have not been placed in context with other N fluxes in the ecosystems under study.Larsen (1980) constructed a schematic N cycle for the boreal forest which incorporated an estimate of nitrogen fixation for S. paschak derived from acetylene reduction data of Crittenden & Kershaw (1979).This suggested that nitrogen fixation by this species <20 kg ha -1 y 1 ) could potentially provide much of the annual N requirement of vascular plants (10-40 kg ha 1 y 1 ) in upland spruce forest.Stereocaulon paschale is an active nitrogen fixer that has the potential to form extensive, almost monospecific, groundcover in lichen-woodlands.Such ecosystems are abundant in north-central Canada (Kershaw, 1977) and parts of Russia (Lavrenko & Sochava, 1956;Karev, 1961) and will make excellent model systems in which to investigate some key questions that to date remain unexplored.For example, by which pathways and in what quantities does N 2 fixed by lichens reach vascular plants?Does the presence of N 2 -fixing lichens influence rooting patterns, and promote the growth, of shrubs and trees?These questions have a direct beating on the primary productivity of lichen-rich ecosystems.Moreover, N 2 -fixing lichens are particularly sensitive to air pollution (e.g.Hallingback & Kellner, 1992) giving added significance to these questions.Species of Peltigera and Nephroma ate like- ly to be among the first lichens to be eliminated from subarctic areas affected by acid rain and there is a pressing need to better understand their ecological role.

Response to pollution
Lichens are renowned for their ability to accumulate pollutants and for their usefulness as indicators of pollution loads.Perhaps of most significance for reindeer and caribou production is the capacity for lichens to intercept and accumulate radionuclides in atmospheric deposits (Nevstrueva et al, 1967;Taylor et al, 1988;Feige et al, 1990;Baskaran et al, 1991;Strandberg, 1997).The food chain concentrating effect has been implicated as the cause of an increased incidence of cancer among some native peoples in the Arctic (Crittenden, 1995).
There is now growing interest in the impact of acid rain in northern latitudes.Acid rain contains elevated concentrations of both acidity and N, principally as nitrate.The concentration of NO, in precipitation has steadily increased in northern latitudes during the past century (Brimblecombe & Stedman, 1982;Rodhe & Rood, 1986;Fischer et al, 1998).Recent studies have shown mat-forming lichens to be sensitive indicators of N and acid deposition.Using Cladonia portentosa in British heathlands as a model system, Hyvarinen & Crittenden (1996;1998b) showed that the ratio K* : Mg 2+ in the thallus apices was highly correlated with H* concentration in precipitation (Fig. 3), and that thallus N concentration was related to total N deposition (Fig. 4).While a preliminary study could not find any evidence that the growth of C. portentosa was impaired by acid rain (Hyvarinen & Crittenden, 1998a) it was often absent from eutrophicated heathlands.This is probably owing to competitive exclusion by mosses and vascular plants, the growth of which is believed to be stimulated by elevated N deposition driving a change from slower to faster growing species.Mat-forming lichens may provide particularly sensitive indicators of atmospheric chemistry because (i) they occur in open situations where they intercept atmospheric deposits directly and largely unmodified by plant canopies, (ii) they are partially isolated from the chemical status of underlying soil by a layer of necromass, and (iii) they are very efficient scavengers of inorganic N (and probably P) in precipitation (Crittenden, 1983;1989;Hyvarinen & Crittenden, 1998c).For example, nitrogen concentration in the apical 5 mm of C. stellaris was measured at sites distributed along a 240 km long transect across the taiga-tundta ecotone in northern Russia and found to gradually decrease from south to north (Fig. 5).This change probably reflects a south to north decrease in N deposition due to either a decline in total precipitation (ranging from c. 1200 mm at the southern end of the transect to c. 620 mm at the northern end) or a pollution gradient, ot both.However, a comparable gradient in nitrate concentration in winter snow pack was not observed (Walker, Crittenden & Young, unpubl.).
Ecosystems at high latitudes are classically Nlimited and potentially responsive to N pollution.This is likely to increase in the future due to such activities as exploitation of oil and gas reserves located deep within the Subarctic.Nitrogen pollution has been implicated as a factor contributing to increased growth rates of northern boreal forest trees (D'Arrigo et al, 1987;Myneni et al, 1997;Nadelhoffer et al, 1999).The effect on lichens will depend on the N load.Species containing cyanobacteria, such as Stereocaulon, Peltigera and Nephroma may be particularly susceptible to increased N and acid deposition although the latter species are not favoured by reindeer or caribou.However, increased vascular plant vigour could result in the competitive exclusion of terricolous lichens in the most affected areas.S0chting (1990; 1995) considers this to be a contributory factor in the decline of Cladonia (subgenus Cladind) spp. in Danish heathlands.In some of the most eutrophicated heathlands in the Netherlands and the UK there has been an incursion of graminoids resulting in competitive exclusion of not only lichens but also ericaceous shrubs: in these areas heathlands are changing into grasslands (Aerts & Heil, 1993;Marrs, 1993).

Concluding remarks
The majority of lichen research is curiosity-driven and takes place in countries without reindeer industries.It attracts minimal financial support and lichenologists are not generally accustomed to having end-users who consider lichens of practical importance.Perhaps it is not surprising, therefore, that lichenology is poorly equipped to address some of the range management issues facing reindeer herding.One way forward is for those working in reindeer range management to identify key questions that they think lichenologists should address.These can be publicised in international fora and will hopefully provide incentive for research on topics that are currently neglected and perhaps viewed as unfashionable.The rate of mass increase in matforming lichens is clearly one such topic.

Fig. 1 .
Fig. 1.Examples of mat-forming lichens, a) Cladia retipora (Australasia), b) Cladonia stellaris, c) Stereocaulon paschale and d) Cetraria cucullata (boreal-arctic zone).Cladia retipora occurs in coastal and alpine heathlands in Australasia.Cladonia stellaris is the supreme example of a mat-forming lichen and possibly one of the world's most abundant fungi; species that are morphologically similar to C. stellaris occur in the Southern Hemisphete e.g. C. pycnoclada, C. leptoclada and C. confusa, which locally form extensive ground cover.

Fig. 4 .
Fig. 4. The relationship between mean values (with ± 1 standard error) of total nitrogen concentration («=5-10) in a horizontal stratum of Cladonia portentosa between 40-50 mm from the thallus apices ([N] baS( .)and wet deposition of NO3 as a function of total N deposition.