Species Level Patterns in 13C and 15N Abundance of Ectomycorrhizal And

Research

SpeciesBlackwell Publishing Ltd. level patterns in 13C and 15N abundance of ectomycorrhizal and saprotrophic fungal sporocarps

Andy F. S. Taylor1, Petra M. Fransson1, Peter Högberg2, Mona N. Högberg2 and Agneta H. Plamboeck3 1Department of Forest Mycology and Pathology, Swedish University of Agricultural Sciences, PO Box 7026, SE-750 07 Uppsala, Sweden; 2Section of Soil Science, Department of Forest Ecology, Swedish University of Agricultural Sciences, S-901 83 Umeå, Sweden. 3Center for Stable Isotope Biogeochemistry, Department of Integrative Biology, University of California, Berkeley, CA 94720, USA.

Abstract

Author for correspondence: • The natural abundance of 13C (δ13C) and 15N (δ15N) of saprotrophic and ecto- Andy Taylor mycorrhizal (ECM) fungi has been investigated on a number of occasions, but the Tel: +46 18 672797. significance of observed differences within and between the two trophic groups Fax: +46 18 673599 Email: [email protected] remains unclear. • Here, we examine the influence of taxonomy, site, host and time upon isotopic Received: 24 February 2003 data from 135 fungal species collected at two forest sites in Sweden. Accepted: 15 May 2003 • Mean δ13C and δ15N values differed significantly between ECM and saprotrophic doi: 10.1046/j.1469-8137.2003.00838.x fungi, with only a small degree of overlap even at the species level. Among ECM fungi, intraspecific variation in δ15N was low compared with interspecific and intergeneric variation. Significant variation due to site, year and host association was found. • At broad scales a number of factors clearly influence δ13C and δ15N values making interpretation problematic. We suggest that values are essentially site-specific within the two trophic groups, but that species-level patterns exist potentially reflecting ecophysiological attributes of species. The species is therefore highlighted as the taxonomic level at which most information may be obtained from fungal δ13C and δ15N data. Key words: fungal diversity, functional groups, nutrient cycling, ectomycorrhizal (ECM) fungi, saprotrophic fungi, stable isotopes. © New Phytologist (2003) 159: 757–774

fungi have the potential to assimilate many of the major Introduction nitrogen (N)- and phosphorus (P)-containing organic Ectomycorrhizal (ECM) fungi, obligate root symbionts of molecules in plant, microbial and animal detritus (Leake most boreal forest trees, and saprotrophic macromycetes & Read, 1997). This may occur either directly via the contribute most to the observed fungal diversity within boreal production of catabolic extracellular enzymes or indirectly forest ecosystems (Bills et al., 1986; Såstad & Jenssen, 1993; via combative interactions with saprotrophic fungi (Lindahl Renvall, 1995; Väre et al., 1996). Traditionally, these fungi et al., 1999). In addition, some fungi formally regarded as have been regarded as two distinct functional groups within saprotrophs are now known to be ECM fungi (see Agerer & ecosystems (Dighton, 1995; Leake & Read, 1997) with Beenken, 1998; Erland & Taylor, 1999; Kõljalg et al., 2000). saprotrophic fungi obtaining carbon (C) and nutrients from Direct in situ observation of fungi is difficult because of the the degradation of organic compounds (Swift et al., 1979) small size of the vegetative structures and the opaque nature and ECM fungi facilitating the uptake of nutrients by their of the growing medium. Consequently, most knowledge con- autotrophic host plants in return for fixed C as photosynthate cerning the involvement of these fungi in ecosystem processes (Smith & Read, 1997). However, over the past two decades is derived from laboratory investigations. One method that the distinction between the two groups has become less well has, in recent years, been investigated as a potential indirect defined. There is now considerable evidence that some ECM method for assessing the functional roles of the fungi in

758 Research

ecosystems is the analysis of the stable isotopes 13C and 15N capabilities, carbon demand and habitat preferences (Tyler, in fungal material. 1985; Leake & Read, 1997; Cairney, 1999). All of these Cycling of C and N through different components of factors could be expected to affect the δ15N and δ13C values ecosystems create small but measurable differences in the of the sporocarps produced by individual ECM species. It is isotope ratios of 13C : 12C and 15N : 14N (Dawson et al., 2002). also well established that there are distinct successions of These differences have been used extensively to investigate saprotrophic fungi colonizing plant debris (Renvall, 1995), with plant ecology and the pathways of C and N cycling through each fungal species or group of species capable of utilizing ecosystems (Farquhar et al., 1989; Fung et al., 1997; Högberg, different chemical components of the plant material (Tanasaki 1997) and this approach has also been used as a tool to et al., 1993). As these components may vary with respect to investigate ECM and saprotrophic fungal ecology (Hobbie 15N (Högberg, 1997) and 13C abundance (Gleixner et al., et al., 1999a, 2001; Högberg et al., 1999b; Henn & Chapela, 1993), it seems inevitable that the 15N and 13C signatures of 2001). Direct analysis of soil mycelia is rarely practical and sporocarps will reflect a differential substrate usage (Gebauer nearly all of these studies have involved determining the & Taylor, 1999). natural abundance of 15N and 13C in the sporocarps of If differences in the 15N and 13C abundance of sporocarps macromycetes. One exception to this is the study by Högberg are a reflection of the physiological diversity among fungal et al. (1996), where the 15N abundance in the mantles of species, then the data may be most informative at the taxonomic beech (Fagus sylvatica) mycorrhizas was determined. Some level of the species. Support for this idea comes from the study general patterns are becoming apparent from these studies. by Högberg et al. (1999b), which demonstrated that by Ectomycorrhizal fungi are generally more depleted in 13C and analysing 13C abundance data from mycorrhizal fungi at the more enriched in 15N than saprotrophic fungi (Hobbie et al., species level, host specificity between trees and their associated 1999b; Högberg et al., 1999b; Kohzu et al., 1999; Henn & ECM fungi could be examined in situ for the first time. Chapela, 2001). Among the saprotrophs, litter fungi are more The present study examines the 15N and 13C natural enriched in 15N than wood decomposers, with both groups abundance in sporocarps of ECM and saprotrophic fungi to enriched relative to their substrates (Gebauer & Taylor, 1999; determine (1) how site, host and time influence 15N and 13C Kohzu et al., 1999). natural abundance and (2) if species-specific patterns exist. In Another general pattern is emerging in which ECM fungi addition, inter-yearly variation was also examined. The data is are considerably enriched in 15N relative to their host plants derived from extensive collections of ECM and saprotrophic (Gebauer & Dietrich, 1993; Högberg et al., 1996, 1999a; fungi at sites in central and northern Sweden. The results Taylor et al., 1997; Michelsen et al., 1998; Hobbie et al., show that potentially important ecological information 1999b). Given that > 95% of the host root tips are usually may be lost when 15N and 13C natural abundance data are colonized by ECM fungi (Dahlberg et al., 1997; Fransson interpreted above the level of the species and that using the et al., 2000; Taylor et al., 2000), most of the N taken up by the data to distinguish between saprotrophic and ECM fungi tree would have to pass through the fungi. It could therefore must be done with caution. be expected that the 15N signatures of the fungi and the host plant would be similar. Fractionation during transfer of N Materials and Methods from the soil through the fungal tissue to the host plant has been suggested as the main reason for the observed differences Sites between the host and fungal 15N abundances (Hobbie et al., 1999b; Högberg et al., 1999a). The study was conducted at Stadsskogen in Uppsala, central In many of the field investigations cited above, fungi are Sweden (59°52′N, 17°13′E, 35 m above sea level) and at often split into large ecological groups (ECM, litter or wood Åheden in the Svartberget Research Forest, 60 km north-west decomposers), within which species are considered to carry of Umeå, northern Sweden (64°14′N, 19°46′E, 175 m above out similar functions within the ecosystem. It therefore sea level). Both forests have 150-yr-old Scots pine (Pinus follows that within a functional group, species are expected sylvestris L.) as an overstorey with Norway spruce (Picea abies to have similar isotope signatures and conclusions concerning (L.) Karst.) approaching the role of a codominant. At the significance of observed values are therefore often made at Stadsskogen, understorey deciduous tree species included these very broad functional levels, regardless of the taxonomic aspen (Populus tremula L.), birch (Betula pendula Roth), diversity of the organisms involved. There is, however, con- willow (Salix caprea L.), and alder (Alnus incana (L.) Moench). siderable evidence, both physiological and ecological, which At Åheden, birch was the only understorey tree species. strongly suggests that this assumption of ecological equiva- lency within fungal functional groups is invalid (Smith & Sampling Read, 1997). Among ECM fungi, much variation exists among species, Sampling was conducted in an area of c. 1 ha at Åheden in even within the same genus, with regard to enzymatic August 1997 and at five plots, each with an area of 0.25 ha at

www.newphytologist.com © New Phytologist (2003) 159: 757–774

Research 759

Stadsskogen between August and October 1997. Between using  and Tukey’s family error (5%) test. Temporal one and eight mature sporocarps from each fungal species effects were examined by calculating mean values for each of were sampled at each site (see Appendix 1). Nomenclature the six sampling periods based on all sporocarps collected at primarily follows that of Hansen & Knudsen (1992, 1997). that time. Twelve ECM species occurred with two or more At Åheden, sampling in 1993, 1995 and 1997 allowed sporocarps on both sites, Stadsskogen and Åheden. These examination of interannual variability (see later). In addition, species were used to make intersite comparisons of δ15N, %N at each site, leaves and current needles of tree species were δ13C, %C-values and C : N ratios, using paired t-tests. sampled (> 10 g dry wt per sample) from branches on the Inter-yearly differences in (δ15N were examined by compar- south side, close to the top of the tree, from up to 10 ing data from sporocarps collected at Åheden in 1993 and specimens per tree species. 1995 (data from Taylor et al., 1997) with the mean value for 1997. This was possible for six species that were collected on three or more occasions in 1997. The 95% confidence interval Preparation and sample analysis was calculated for the mean value for each of these six species The sporocarps and foliar samples were dried (700C, 24 h) and any value from 1993 or 1995 which lay outside this and then ground in a ball mill. Samples were analysed for interval was therefore considered significantly different from 15N and 13C abundance using an on-line continuous flow the mean at P = 0.05. CN analyser coupled to an isotope ratio mass spectrometer (Ohlsson & Wallmark, 1999). Results are expressed in the Results standard notation (δ13C and δ15N) in parts per thousand (‰) relative to the international standards V-PDB and Comparison of functional groups (ECM and atmospheric N , where δ13C or δ15N = ((R /R ) − 2 sample standard saprotrophic fungi) 1] × 1000, and R is the molar ratio 13C : 12C or 15N : 14N. The standard deviation based on the analysis of replicated A total of 135 fungal species (118 ECM and 17 saprotrophic) samples was 0.15‰ and 0.20‰ for 13C and 15N, respectively. representing 25 ECM and 15 saprotrophic genera were analysed in the study (Appendix 1). In general, the mean values for (δ15N, %N δ13C, and %C values and C : N ratios Statistical analysis differed significantly between saprotrophic and ECM fungi, Many species were represented by more than one collection irrespective of site (Table 1). The general pattern was for (a sporocarp) and this is a potential source of bias when ECM fungi to be more enriched in 15N, but more depleted in calculating mean values for different groups or categories of 13C, and have a higher percentage of carbon than saprotrophic fungi. It was therefore necessary to calculate weighted means fungi. At Stadsskogen, the %N of ECM sporocarps was that took this into account. Weighted means of δ15N, %N, significantly lower than in saprotrophs, but at Åheden, the δ13C, %C and C : N-values were calculated separately for reverse pattern was observed. The %N varied considerably saprotrophic and mycorrhizal fungi. A mean value for each more than %C, resulting in sporocarp C : N ratios being group was obtained by calculating the sum of the mean values largely determined by the percentage of N they contained for the different fungal species in the group and then by (Table 1). The significant difference observed in the C : N dividing this by the number of species in the group. The ratio between the two life forms reflects the changing N status estimated variance of the group mean was the sum of the of the two groups (Table 1). variances of the species means, divided by the square root of Plotting δ15N against the δ13C values for the individual number of species in the group. Group means were then species separated most saprotrophic from ECM fungi (Fig. 1). compared using Tukey’s test at the 5%, 1%, and 0.1% levels. However, at Stadsskogen there was some overlap between the Most analyses were restricted to Stadsskogen, from which two groups. Adding a third axis representing %N, %C or the we obtained the more complete data set. For the ECM fungi, C : N ratio of the sporocarps did not significantly increase the (δ15N, %N, (δ13C, %C and C : N data were compared at the distinction between the groups (data not shown). the family and generic taxonomic levels. The weighted means of the nine ECM families represented by more than three Comparison of ECM fungi at different taxonomic levels species and intergeneric differences were compared using one way analysis of variance () and Tukey’s family error The data for ECM fungi analysed at different taxonomic levels (5%) test. Possible differences related to host specificity on the is presented in the following increasing order of resolution: major tree species (Pinus, Picea and Betula) were compared in family, genus and species. Fourteen ECM families were the same way. The δ13C data has been presented in detail represented in the collections from Stadsskogen with nine of previously (Högberg et al., 1999b). these represented by three or more species. There were highly Temporal effects due to sampling date and differences significant differences among these nine families with respect between the five plots sampled in Stadsskogen were also tested to %N, δ15N, %C δ13C values and C : N ratios (Table 2).

Table 1 Mean values for percentage nitrogen (%N), δ15N, percentage carbon (%C), δ13C and C : N ratios in sporocarps of ectomycorrhizal (ECM) and saprotrophic fungi collected from two forests in Sweden (Åheden, northern Sweden; Stadsskogen, central Sweden)

Åheden ECM Saprotrophic Stadsskogen ECM Saprotrophic fungi (n = 28) (n = 5) P1 fungi (n = 109) (n = 14) P1

δ15N (‰) 7.8a2 −0.8 < 0.001 5.8b 0.8 < 0.001 δ13C (‰) −25.8 −23.4 < 0.001 −25.8 −23.2 < 0.001 %N 3.9 3.1a 0.05 > P > 0.01 3.9 5.8b < 0.001 % C 43.0 42.2 ns 43.5 41.7 < 0.001 C/N ratio 11.9 15.1 0.01 > P > 0.001 11.5 9.2 < 0.001

1Significance of within site differences between fungal life strategies, analysed using students t-test; ns, not significant. 2Between-site comparisons of mean values for each life form; means with different letters are significantly different at P = 0.05.

Table 2 Mean values ± SE for percentage nitrogen (%N), δ15N, percentage carbon (%C), δ13C and C : N ratios in sporocarps of ectomycorrhizal fungi from nine families, collected in a mixed boreal forest at Stadsskogen, Uppsala, central Sweden

Family %N1 δ15N (‰) %C δ13C (‰) C : N

Amanitaceae 4.3 ± 0.2 a,c,d 4.1 ± 1.1 a,b 44.1 ± 0.6 a,b,c −25.5 ± 0.3 a,b 10.3 ± 0.4 a,b Boletaceae 4.9 ± 0.3 c,d 8.7 ± 1.3 b 42.4 ± 0.3 d −25.8 ± 0.5 b 9.1 ± 0.6 a Gomphidiaceae 3.4 ± 0.2 a 7.7 ± 1.5 a,b 43.5 ± 0.4 a,d,e −25.1 ± 0.2 a,b 13.0 ± 0.9 b Hygrophoraceae 3.9 ± 0.4 a,c 2.6 ± 2.3 a,b 42.5 ± 0.3 c,d −26.2 ± 1.0 b 11.1 ± 1.2 a Tricholomataceae 3.8 ± 0.3 a,b 9.0 ± 1.5 b 42.8 ± 0.4 c,d −25.6 ± 0.4 b 11.6 ± 0.9 a,b Cortinariaceae 4.1 ± 0.1 a,e 5.6 ± 0.4 a,b 42.7 ± 0.1 d −26.1 ± 0.2 b 10.8 ± 0.3 a,b Russulaceae 3.8 ± 0.1 a 4.3 ± 0.5 a 44.3 ± 0.2 a,e −25.8 ± 0.2 b 11.9 ± 0.3 b Thelephoraceae 5.1 ± 0.4 b,c,e 9.7 ± 0.8 a,b 46.3 ± 0.3 b,f −22.8 ± 0.2 a 9.4 ± 0.9 a,b Cantharellaceae 3.1 ± 0.1 a 3.4 ± 1.1 a,b 43.9 ± 0.3 a,d,f −25.5 ± 0.4 a,b 14.2 ± 0.5 b F (d.f. 8,95). 5.30 5.38 12.38 3.04 4.96 P < 0.001 < 0.001 < 0.001 < 0.0042 < 0.001

1Within columns mean values sharing the same letter are not signiﬁcantly different. 2Excluding the data from Thelephoraceae gives a nonsigniﬁcant result (F = 0.78, P = 0.603).

A total of 24 genera were collected at Stadsskogen, of which eight were represented by three or more species. The mean values for %N, δ15N, %C δ13C and C : N ratios varied considerably among these eight genera (Fig. 2). The δ13C values clearly separated those genera that were either nearly or entirely represented by species forming specific associations with Pinus (Suillus) or Betula (Leccinum). The 13C values of other genera that were represented by a range of specific and nonspecific species did not differ. The δ15N-values also varied considerably among the genera, even between the closely related genera Russula and Lactarius (Fig. 2). As mentioned above, the C : N ratio of the sporocarps was largely determined by the percentage of N they contained and this relationship was also evident at the generic level (R 2 = 0.962; P < 0.001). With this exception, no other two parameters were significantly correlated at the generic level.

Fig. 1 Natural 15N and 13C abundance in sporocarps of ectomycorrhizal (ECM) (open circles) and saprotrophic (triangles) fungal species collected from mixed boreal forests at (a) Åheden, northern Sweden, and (b) Stadsskogen, central Sweden. Data points represent mean values (n = 1–8). (*, Chalciporus piperatus, traditionally considered ectomycorrhizal but the δ13C value clearly separates it from the other ECM fungal species).

www.newphytologist.com © New Phytologist (2003) 159: 757–774 Research 761

Fig. 2 Inter-generic variation (mean ± SE) in the natural abundance of 15N and 13C in sporocarps of ectomycorrhizal fungi collected in a mixed boreal forest at Stadsskogen, central Sweden. Numbers in brackets represent the number of species analysed within each genus.

Fig. 4 Intrageneric variation (mean ± SE) in the natural abundance of 15N and 13C in sporocarps of the ectomycorrhizal genera (a) Amanita, (b) Cantharellus, (c) Suillus and (d) Russula collected in a mixed boreal forest at Stadsskogen, central Sweden. Numbers in brackets represent the number of collections analysed within each species.

Fig. 3 Intra-generic variation (mean± SE) in the natural abundance of 15N and 13C in sporocarps of the ectomycorrhizal fungal genus Lactarius collected in a mixed boreal forest at Stadsskogen, In particular, within the genus Suillus there was a striking central Sweden. Host preference in brackets (Gen., generalist). difference between S. variegatus and the two other species 1, L. badiosanguineus (Picea); 2. L. camphoratus (Gen.): 3, examined (Fig. 4c). L. deliciosus (Pinus); 4, L. deterrimus (Picea); 5, L. fuliginosus (Gen.); 6, L. glyciosmus (Betula); 7, L. helvus (Gen.); 8, L. mitissimus (Gen.); 9, L. musteus (Pinus); 10, L. necator (Gen.); 11, L. obscuratus Spatial and temporal effects (Alnus); 12, L. repraesentaneus (Gen.); 13, L. rufus (Gen.); 14, L. scrobiculatus (Picea); 15, L. theiogalus (Gen.); At Stadsskogen, there was a significant difference in the mean 16, L. torminosus (Betula); 17, L. trivialis (Gen.); 18, (δ15N value of ECM fungi (F = 3.14, P = 0.011) among L. uvidus (Betula); and 19, L. vietus (Betula). plots. The sporocarps from one area (δ15N = 7.1) were enriched by as much as 2.1‰ relative to that in three other areas (δ15N = 5.0–5.1). There were no significant differences Intrageneric variation was high: in most cases where genera in the %N, δ13C and %C values or C : N ratio among plots. were represented by two or more species, there were significant There was some evidence that sporocarp %N (F = 2.06, differences in both δ15N and δ13C values between species. For P = 0.070) and δ13C (F = 1.87, P = 0.099) were both affected example, within the genus Lactarius mean δ15N values for by sampling date. The %N value did not show any particular individual species ranged from 1.7 to 8.5‰ (Fig. 3), with the pattern over the sampling period, but δ13C showed a clear 19 species represented spread more or less evenly across this trend to decrease as the fruiting season progressed (Fig. 5). range. The δ13C values also showed considerable variation The δ15N values of the ECM fungi at Åheden were sig- among species, ranging from −27.1 to −23.5‰. However, nificantly higher than those at Stadsskogen (Table 1). An there was greater overlap in δ13C values among species than analysis of 12 ECM species that were represented by two with the δ15N values. Other genera also showed marked or more collections at both sites revealed that with the variation among species with regard to δ15N values (Fig. 4). exception of Amanita muscaria, sporocarps were generally

enriched in 15N at Åheden (Fig. 6). The %N of sapro- value on at least one of the two previous collecting periods trophic fungi was, however, higher at Stadsskogen than at (Table 3). Åheden. No clear pattern was found with respect to %C or δ13C values. The 15N and 13C natural abundance of tree hosts and It was possible to construct confidence intervals for the host-specific fungi mean δ15N values of six species collected at Åheden in 1997. A comparison between these and data from collec- At Åheden, the two codominant tree species, pine and spruce, tions made at Åheden in 1993 and 1995 (data from Taylor had similar 13C values (Table 4), with both values significantly et al., 1997) demonstrated that values from five out of the higher than those of birch. At Stadsskogen, δ13C increased in six species were significantly different from the 1997 mean the following order: birch < spruce < pine. The other understorey tree species at Stadsskogen had values very similar to those of the birch (data not shown). The δ13C values of the fungi and their relationship to host values have been presented previously (Högberg et al., 1999b). The 15N values of pine, spruce and birch differed little within site and only birch differed between sites (Table 4, Fig. 7). There was a clear difference between sites with respect to host-specific fungi (Fig. 7), with δ15N-values significantly higher at Åheden (paired t-test T = 4.63, P = 0.04). The difference between fungal and plant δ15N was also greater at Åheden (difference in δ15N = 3.2 ± Fig. 5 Mean natural 13C abundance in sporocarps of ectomycorrhizal 0.1‰, paired t-test T = 37.04, P = 0.0007). On average, the fungi collected in a mixed boreal forest at Stadsskogen, central δ15 Sweden. Sporocarps were collected at six sampling periods, approx. differences in N between host and host-specific fungi were 2 wk apart from August to October 1999. 8.4‰ and 11.7‰ at Stadsskogen and Åheden, respectively.

Fig. 6 Natural 15N abundance in the sporocarps of 12 species of ECM fungi collected from two mixed boreal forests at Stadsskogen, central Sweden (ﬁlled columns) and Åheden, northern Sweden (tinted columns). A. mu., Amanita muscaria; Ch. r. – Chroogomphus rutilus; Co. a., Cortinarius armillatus; Co. l., Cortinarius laniger; Co. m., Cortinarius malachius; D. se., Dermocybe semisanguineus; La. r., Lactarius rufus; Le. s., Leccinum scabrum; Le. vs., Leccinum versipelle; R. c., Rozites caperata; S. va., Suillus variegatus; T. ﬂ., Tricholoma flavovirens.

Table 3 Inter-annual variation in the δ15N values in sporocarps of ectomycorrhizal fungi collected in a mixed boreal forest at Åheden, northern Sweden

Mean value 95% CI1 Species n for 1997 for 1997 1993 Value2,3 1995 Value2,3

Chroogomphus rutilus 4 3.8 1.4–6.2 0.9 6.2 Cortinarius laniger 6 10.1 8.7–11.5 12.7 9.7 Dermocybe semisanguinea 3 7.4 4.3–10.5 5.9 7.5 Lactarius rufus 6 4.0 3.0–4.9 1.8 3.2 Suillus variegatus 6 6.0 4.9–7.2 4.8 4.6 Tricholoma flavovirens 5 9.9 8.2–11.7 10.6 7.6

195% Conﬁdence interval for the mean 15N abundance in sporocarps collected in 1997. 2Values in bold type are signiﬁcantly different from the 1997 mean values. 3Data from Taylor et al. (1997).

www.newphytologist.com © New Phytologist (2003) 159: 757–774 Research 763

Table 4 Mean values for percentage nitrogen (%N), δ15N, percentage carbon (%C), δ13C and C : N ratios in current tree foliage collected from two forests in Sweden (Åheden, northern Sweden; Stadsskogen, central Sweden)

Site Tree species n %N δ15N (‰) %C δ13C (‰) C : N

Åheden Pinus sylvestris 9 1.5 ± 0.1 a1 −2.9 ± 0.2 50.6 ± 0.1 −27.0 ± 0.3 a 32.8 Picea abies 9 1.4 ± 0.1 a −3.8 ± 0.2 52.1 ± 0.2 −27.4 ± 0.3 a 37.2 Betula pendula 9 2.3 ± 0.1 b −4.1 ± 0.82 49.5 ± 0.2 −29.6 ± 0.8 b 21.9 Stadsskogen Pinus sylvestris 10 1.6 ± 0.1 a −2.8 ± 0.3 50.9 ± 0.2 −26.8 ± 0.2 a 32.4 Picea abies 10 1.4 ± 0.1 a −3.5 ± 0.3 50.2 ± 0.1 −27.9 ± 0.3 b 36.1 Betula pendula 10 2.2 ± 0.1 b −2.2 ± 0.52 50.3 ± 0.4 −29.7 ± 0.2 c 22.7

1Within site, means not sharing the same letter are signiﬁcantly different at P = 0.05. 2The δ15N values for birch differed signiﬁcantly between sites (T = −2.36, P = 0.033).

Fig. 7 Natural 15N abundance in host plants and in the sporocarps of host-speciﬁc ectomycorrhizal fungi at Stadsskogen, central Sweden (ﬁlled columns) and Åheden, northern Sweden (tinted columns).

the saprotrophic fungi. New data (unpublished) obtained Discussion from three other Agaricus species (A. aestivalis, A. arvensis and A. bitorquis) also show consistently high δ15N values Differences between trophic groups (7–13.2‰). The high δ15N values may reflect the high %N A number of recent studies have suggested that ECM and contents of these species. saprotrophic fungi may be separated on the basis of δ15N, A small number of ECM species produced sporocarps with and δ13C values (Hobbie et al., 1999b; 2001; Högberg et al., either low δ15N values or high δ13C, values placing them in 1999b; Kohzu et al., 1999; Henn & Chapela, 2001). A the saprotrophic group. Some presumed ECM species, most comparison between the mean values for the two groups notably Chalciporus piperatus, Hydnellum ferrugineum, Hyd- within the present study strongly supports this idea. However, nellum peckii and Phellodon niger, overlapped strongly with when viewed at the species level, only at Åheden were species saprotrophic δ13C values. As pointed out previously (Högberg from each group clearly separated from each other (Fig. 1). An et al., 1999b), the mycorrhizal status of C. piperatus is un- analysis of data from all collections from Stadsskogen (327 confirmed, but according to its isotopic signature it seems ECM and 21 saprotrophic sporocarps) showed some overlap likely that this taxon is a saprotroph. The three other species between ECM and saprotrophic fungi. In particular, the δ15N are known to form mycorrhizas (Agerer, 1986–98). However, values from two species of terricolous (growing on forest floor) they differ from other ECM species included in this study saprotrophs (Clitocybe clavipes and Agaricus haemorrhoideus) because their sporocarps survive for a much longer period were high, placing them well within the ECM group. (> 1 month) than other species and the mycorrhizas associated However, their δ13C values clearly associated them with with the sporocarps appear to be degraded. Ectomycorrhizas

formed by these species that are not associated with sporocarps in 15N (Högberg, 1997). It seems likely that the high values are, however, morphologically typical of other ECM species for Hebeloma reported by Kohzu et al. (1999) are a result of (A. Taylor, pers. obs.). The high δ13C values and the state of this process. the sporocarp-associated mycorrhizal tips suggest that, during As mentioned above, C sources differ between saprotrophic sporocarp formation, the mutualistic balance favours the fungus. and ECM fungi. By contrast to this the mycelia of both The reduced distinction between ECM and saprotrophic groups are, with few exceptions, likely to forage in the same trophic groups at Stadsskogen may, at least in part, be due to substrates for N. In the absence of fractionation during N the much greater number of species sampled compared with metabolism this sharing of resources should mean considerable Åheden. It is possible that further collections, particularly of overlap in δ15N-values between the two groups. However, it terricolous saprotrophs, from the sites may further increase is clear from the present study and previous works that mean the degree of overlap. δ15N-values for the two groups may differ significantly Saprotrophic and ECM fungi may be distinguished on the (Hobbie et al., 1999b). Fractionation against the heavier 15N basis of how they obtain their C. Ectomycorrhizal fungi isotope during the transfer of N from ECM fungi to the host depend primarily on current photosynthate from host plants plant has been proposed to explain this difference (Högberg (Söderström & Read, 1987; Högberg et al., 2001) and sapro- et al., 1996, 1999a; Taylor et al., 1997; Hobbie et al., 1999b). trophic fungi degrade organic matter (Swift et al., 1979). In addition, this fractionation step is thought to explain the These C sources are likely to differ in 13C content (Benner observed difference between host and ECM δ15N-values. et al., 1987; Nadelhoffer & Fry, 1988; Gleixner et al., 1993), Fractionation against the heavier isotope during the transfer of providing a basis for the observed differences in 13C signa- N to the host plant has recently been confirmed in laboratory tures in fungi (Henn & Chapela, 2001). The differentiation studies (Högberg et al., 1999a; Kohzu et al., 2000; Emmerton between the two functional groups (termed the EM/SAP et al., 2001). Thus, the δ15N of saprotrophic fungi, which do divide, Henn & Chapela, 2001) has been used to infer ECM not transfer N taken up any further, should reflect the isotopic or saprotrophic lifestyles for fungal taxa of unknown trophic signature of the source, while the δ15N of ECM fungi is status. dependent on both source signature and the efficiency of Henn & Chapela (2001) performed a field study in transfer of N to their host plant (Högberg et al., 1999a; California and a meta-analysis of three earlier studies and Hobbie et al., 2000). However, unlike the ECM fungi, the N found strong evidence for a distinction between ECM and concentration and δ15N of saprotrophs growing on litter at saprotrophic fungi in isotopic signatures. However, they Stadsskogen is strongly correlated (r = 0.966, P < 0.001, n = emphasized that the accuracy of distinguishing between the 9). This correlation was also seen in the saprotrophs examined groups deteriorated when data from geographically different by Hobbie et al. (2001). This relationship between N concen- areas were compared. In addition, doubt was raised over tration and δ15N would probably not be found if the signature studies where trophic status was assigned to fungal taxa on the of the N source used was the single determinant of δ15N of basis of single sporocarps. saprotrophic fungi. Kohzu et al. (1999) reported δ15N and δ13C values from a large number of ECM, wood-decomposing and litter- Taxonomic level patterns decomposing fungi from a range of forest types in Japan and Malaysia. In the Japanese material they found considerable One striking feature of the present study is that irrespective of overlap in the δ15N-values from all three groups of fungi. the taxonomic level at which the ECM data was analysed However, some of this overlap may be explained by the (family, genus or species) there were significant differences inclusion of several fungi of dubious mycorrhizal status within within taxonomic levels. This contrasts with Kohzu et al. (1999) the ECM category (e.g. Entoloma spp., Catathelasma sp. and who found few significant differences at any taxonomic level. Lyophyllum sp.). All three of these taxa have high δ13C values They, however, included material from a wide range of compared with the great majority of the rest of the ECM habitats and sites, making comparisons difficult owing to species. The δ15N mean value for the ECM fungi included in site effects. At Stadsskogen, mean δ13C values at family and the study by Kohzu et al. (1999) was 5.5‰, which is very generic levels were highly influenced by the proportion of similar to that recorded in the present study. The range of host-specific species found at that level. We have reported δ15N values, −6 to +22‰, in ECM species recorded by Kohzu previously that the stratification of the trees at this site results et al. (1999) was even greater than that found in the present in a clear separation between the δ13C values of the foliage of study (−3.6 to +19.4‰). Two Hebeloma species accounted for the mycorrhizal host plants pine, spruce and birch (Table 4) the very high values in their study; both of these are known to (Högberg et al., 1999b). Where fungal genera or families be associated with the buried remains of mammals or faeces contain a high proportion of pine- or birch-specific fungi, this (Soponsathien, 1998; Yamanaka, 1999). Decomposition of separates them from those groups where the majority of the protein rich substrates is often accompanied by ammonia fungal species are nonspecific. Although nonspecific fungi volatilization that can leave the remaining N highly enriched are likely to receive C from a number of hosts, they appear to

www.newphytologist.com © New Phytologist (2003) 159: 757–774 Research 765 receive most of the C from the dominant overstorey pine trees according to the few data that are available (Nadelhoffer & (Högberg et al., 1999b). Fry, 1994; Koba et al., 1998; Hobbie et al., 1999a). The higher By contrast to the significant differences found at higher δ15N-values of the fungi at Åheden can also explain the taxonomic levels, the variation at species level was low and significantly greater difference between host plants and consistent within and between sites. In addition, the species- specific ECM fungi at Åheden compared with Stadsskogen. specific patterns in δ15N values found in species common to The decrease in δ13C values of sporocarps over the vegeta- both study sites suggests that there is an ecophysiological tion season that was found may be explained by changes in explanation for the values observed. This assertion is supported weather related to season. The 13C signal of the carbon that is by the work of Lilleskov et al. (2002) who demonstrated that fixed by the leaves of a plant is influenced by a number of species producing sporocarps with high δ15N-values had factors, including soil moisture, air humidity, temperature greater capacity to utilize organic N sources than those with and radiation (Farquhar et al., 1989, Dawson et al., 2002). As lower values. We have already demonstrated that the 13C autumn progresses temperature and radiation decrease and abundance of host-specific fungi is influenced by the host precipitation often increases. The isotopic discrimination 13 signature (Högberg et al., 1999b). The situation with N is against CO2 during plant photosynthesis is strong but more complex. In addition to a wider range of source signatures variable and can be explained by differing internal CO2 than C, there are a number of steps where fractionation for partial pressures in the leaf under different conditions. The or against 15N may occur. As already stated the difference discrimination is stronger at low rates of photosynthesis, which between host and ECM fungal signatures is thought to be typically occur at low temperatures and low light intensity, primarily due to fractionation against 15N during the transfer and weaker when plants close their stomata during drought of N from the fungus to the host. If this theory is correct, then stress (Farquhar et al., 1989). It has also been shown that δ13C species differences in δ15N will, to some extent, reflect the in needles and wood differs among years due to interannual efficiency of N transfer to the host. Efficiency is used here in differences in weather (Garten & Taylor, 1992; Leavitt, a phytocentric perspective to indicate the fraction of N taken 1993). Furthermore, Pate & Arthur (1998) found that the up by the fungus that is transferred to the host. The greater the 13C abundance of carbon in phloem sap also had seasonal proportion of N that is taken up and transferred, the higher fluctuations, which were reflected about a month later in the the δ15N value of the fungus will be and hence the greater the insoluble carbon of recently formed xylem in the trunk. Thus, difference between the host and the fungus. Some support we expect that the lower δ13C in sporocarps at the end of the comes for this idea from the work of Gorissen & Kuyper fruiting season reflect low temperatures, moist soil and air and (2000) who showed that Suillus spp. supplied more N to the low irradiance. host plant than Laccaria. These genera have high and low 15N values, respectively. Conclusions The use of stable isotopes to investigate the cryptic nature of Spatial and temporal patterns fungal ecology is a potentially powerful tool. However, the We detected both spatial and temporal patterns in the data data and analyses outlined in this paper demonstrate that the set in which spatial variation ranged from local to a regional collection and subsequent interpretation of field data should scale and temporal variation ranged from a seasonal to a yearly be made with caution. Site-related effects may alter both the scale. The differences in 15N values between collection areas in position and the width of the divide between saprotrophic Stadsskogen and between the sites may reflect differences in and ECM fungi. For example, N deposition can significantly N availability, N source and taxonomic composition of the influence the δ15N signatures of forest organisms (Gebauer fungal community sampled. At Åheden (the northern site) et al., 1994; Bauer et al., 2000). Despite this need for caution, low decomposition rates have resulted in an extensive an analysis of δ13C and δ15N data from a site, based on a range accumulation of organic matter at the soil surface. Available of fungal species, can give a good indication of the trophic mineral N levels are low, NO3-N is not detectable in most of status of a species. the soil layers while NH4-N occurs at low concentrations The results highlight the importance of the taxonomic (Persson et al., 2000). At more southern sites, such as composition of the ECM community in determining the Stadsskogen, the mineral N-availability is higher (Persson & overall δ13C and δ15N-values recorded for a site. It is clear that Wirén, 1993). The higher δ15N of sporocarps observed at the significant differences in δ15N-values can be found at both the northern site may, at least in part, reflect a greater use of family and the generic level. However, the large variation in organic N-sources because soil mineral N is close to zero. the data at these higher levels of resolution, at least with Lilleskov et al. (2002) found that there was a strong relationship respect to δ15N values, imply that much of the potentially between ECM species producing sporocarps with high δ15N- useful information may be lost when data are considered above values and their ability to use organic N sources. Organic N- the species level. The observed species-level patterns suggest sources are enriched in 15N compared with mineral-N sources that if there is an ecological functional basis to the measured

values then it is at the level of the species that the data should optimal fertilisation upon a Norway spruce ectomycorrhizal community. be interpreted and the level at which most information may Tree Physiology 20: 599–606. be obtained. Lilleskov et al. (2002) have already demonstrated Fung I, Field CB, Berry JA, Thompson MV, Randerson JT, Malmström CM, Vitousek PM, Collatz GJ, Sellers PJ, Randall DA, δ15 that some ECM species with high N values were able to use Denning AS, Badeck F, John J. 1997. Carbon 13 exchanges between the organic sources of N to a greater degree than species with atmosphere and biosphere. Global Biogeochemical Cycles 11: 507–533. lower values. More studies of this type are needed to verify this Garten CT Jr, Taylor GE Jr. 1992. Foliar δ13C within a temperate deciduous observation. forest: spatial, temporal, and species sources of variation. Oecologia 90: 1–7. Gebauer G, Dietrich P. 1993. Nitrogen isotope ratios in different Acknowledgements compartments of a mixed stand of spruce, larch and beech trees and of understorey vegetation including fungi. Isotopenpraxis 29: 35–44. We thank Sven-Gunnar Ryman (Uppsala Museum) for help Gebauer G, Taylor AFS. 1999. N-15 natural abundance in fruit bodies with identification of sporocarp material. This research was of different functional groups of fungi in relation to substrate utilization. supported by grants from the Swedish Natural Sciences New Phytologist 142: 93–101. Gebauer G, Giesemann A, Schulze ED, Jager HJ. 1994. Isotope ratios and Research Council and the Swedish Council for Forestry and concentrations of sulfur and nitrogen in needles and soils of Picea abies Agricultural Research to P. H. stands as influenced by atmospheric deposition of sulfur and nitrogen compounds. Plant and Soil 164: 267–281. Gleixner G, Danier HJ, Werner RA, Schmidt HL. 1993. Correlations References between the C-13 content of primary and secondary plant-products in different cell compartments and that in decomposing Basidiomycetes. Agerer R. 1986–98. Colour atlas of ectomycorrhizae. Schwäbisch-Gmünd, Plant Physiology 102: 1287–1290. Germany: Einhorn-Verlag. Gorissen A, Kuyper TW. 2000. Fungal species-specific responses of

Agerer R, Beenken L. 1998. Lyophyllum decastes (Fr.) Sing. + Quercus robur ectomycorrhizal Scots pine (Pinus sylvestris) to elevated CO2. L. In: Agerer R, ed. Descriptions of ectomycorrhizae 3. Schwäbisch-Gmund, New Phytologist 146: 163–168. Germany: Einhorn-Verlag, 43.47. Hansen L, Knudsen H. 1992. Nordic Macromycetes, Vol. 2: Polyporales, Bauer GA, Andersen B, Gebauer G, Harrison T, Högberg P, Högbom L, Boletales, Agaricales, Russulales. Copenhagen, Denmark: Nordsvamp. Taylor AFS, Novak M, Harkness D, Schulze E-D, Persson T. 2000. Hansen L, Knudsen H. 1997. Nordic Macromycetes, Vol. 3: Heterobasidioid, Biotic and abiotic controls over ecosystem cycling of stable natural Aphyllophoroid and Gasteromycetoid Basidiomycetes. Copenhagen, nitrogen, carbon and sulphur isotopes. In: Schulze E-D, ed. Carbon Denmark: Nordsvamp. and nitrogen cycling in European forest ecosystems. Berlin Heidelberg, Henn MR, Chapela IH. 2001. Ecophysiology of C-13 and N-15 isotopic Germany: Springer Verlag, 189–214. fractionation in forest fungi and the roots of the saprotrophic-mycorrhizal Benner R, Fogel ML, Sprague EK, Hodson RE. 1987. Depletion of 13C in divide. Oecologia 128: 480–487. lignin and its implications for stable carbon isotope studies. Nature 329: Hobbie EA, Macko SA, Shugart HH. 1999a. Patterns in N dynamics and 708–710. N isotopes during primary succession in Glacier Bay. Alaska. Chemical Bills GF, Holtzmann GI, Millar OK Jr. 1986. Comparison of Geology 152: 3–11. ectomycorrhizal-basidiomycete communities in red spruce versus Hobbie EA, Macko SA, Shugart HH. 1999b. Insights into nitrogen and northern hardwood forests of West Virginia. Canadian Journal of Botany carbon dynamics of ectomycorrhizal and saprotrophic fungi from 64: 760–768. isotopic evidence. Oecologia 118: 353–360. Cairney JWG. 1999. Intraspecific physiological variation: implications for Hobbie EA, Macko SA, Williams M. 2000. Correlations between foliar understanding functional diversity in ectomycorrhizal fungi. Mycorrhiza 9: delta N-15 and nitrogen concentrations may indicate plant–mycorrhizal 125–135. interactions. Oecologia 122: 273–283. Dahlberg A, Jonsson L, Nylund JE. 1997. Species diversity and distribution Hobbie EA, Weber NS, Trappe JM. 2001. Mycorrhizal vs saprotrophic of biomass above and below ground among ectomycorrhizal fungi in an status of fungi: the isotopic evidence. New Phytologist 150: 601–610. old-growth Norway spruce forest in south Sweden. Canadian Journal of Högberg P. 1997. Tansley Review, No. 95 – N-15 natural abundance in Botany 75: 1323–1335. soil-plant systems. New Phytologist 137: 179–203. Dawson T, Mambelli S, Plamboeck AH, Templer PH, Tu KP. 2002. Stable Högberg P, Högbom L, Schinkel H, Högberg M, Johannisson C, isotopes in plant ecology. Annual Review of Ecology and Systematics 33: Wallmark H. 1996. 15N abundance of surface soils, roots and 507–559. mycorrhizas in profiles of European forest soils. Oecologia 108: Dighton J. 1995. Nutrient cycling in different terrestrial ecosystems in 207–214. relation to fungi. Canadian Journal of Botany 73: 1349–1360. Högberg P, Högberg MN, Quist ME, Ekblad A, Nasholm T. 1999a. Emmerton KS, Callaghan TV, Jones HE, Leake JR, Michelsen A, Nitrogen isotope fractionation during nitrogen uptake by ectomycorrhizal Read DJ. 2001. Assimilation and isotopic fractionation of nitrogen and nonmycorrhizal Pinus sylvestris. New Phytologist 142: 569–576. by mycorrhizal and nonmycorrhizal subarctic plants. New Phytologist Högberg P, Plamboeck AH, Taylor AFS, Fransson PMA. 1999b. Natural 151: 513–524. C-13 abundance reveals trophic status of fungi and host-origin of carbon Erland S, Taylor AFS. 1999. Resupinate ectomycorrhizal fungi, with in mycorrhizal fungi in mixed forests. Proceedings of the National Academy emphasis on the genera Amphinema, Piloderma, Pseudotomentella, of Sciences,USA 96: 8534–8539. Tomentella and Tylospora. In: Cairney JWG, Chambers S, eds. Högberg P, Nordgren A, Buchmann N, Taylor AFS, Ekblad A, Ectomycorrhizal fungi: key genera in profile. Berlin, Germany: Högberg MN, Nyberg G, Ottosson-Löfvenius M, Read DJ. 2001. Springer-Verlag, 346–363. Large-scale forest girdling experiment demonstrates that current Farquhar GD, Ehleringer JR, Hubick KT. 1989. Carbon isotope photosynthesis drives soil respiration. Nature 411: 789–792. discrimination and photosynthesis. Annual Review of Physiology Plant Koba K, Tokuchi N, Yoshioka T, Hobbie EA, Iwatsubo G. 1998. Natural and Molecular Biology 40: 503–537. abundance of nitrogen-15 in a forest soil. Soil Science Society of America Fransson PMA, Taylor AFS, Finlay RD. 2000. Effects of continuous Journal 62: 778–781.

www.newphytologist.com © New Phytologist (2003) 159: 757–774 Research 767

Kohzu A, Yoshioka T, Ando T, Takahashi M, Koba K, Wada E. 1999. Persson T, van Oene H, Harrison AF, Karlsson PS, Bauer GA, Cerny J, Natural C-13 and N-15 abundance of field-collected fungi and their Couteaux M-M, Dambrine E, Högberg P, Kjøller A, Matteucci G, ecological implications. New Phytologist 144: 323–330. Rudebeck A, Schulze E-D, Paces T. 2000. Experimental Sites in the Kohzu A, Tateishi T, Yamada A, Koba K, Wada E. 2000. Nitrogen isotope NIPHYS/CANIF Project. In: Schulze E-D, ed. Carbon and nitrogen cycling fractionation during nitrogen transport from ectomycorrhizal fungi, in European forest ecosystems. Ecological studies, Vol. 142. Berlin Heidelberg, Suillus granulatus, to the host plant, Pinus densiflora. Soil Science New York: Springer, 14–46. and Plant Nutrition 46: 733–739. Persson T, Wiren A. 1993. Effects of experimental acidification on C and N Kõljalg U, Dahlberg A, Taylor AFS, Larsson E, Hallenberg N, Stenlid J, mineralization in forest soils. Agriculture Ecosystems and Environment 47: Larsson K-H, Fransson PM, Kårén O, Jonsson L. 2000. Diversity and 159–174. abundance of resupinate thelepheroid fungi as ectomycorrhizal symbionts Renvall P. 1995. Community structure and dynamics of wood-rotting in Swedish boreal forests. Molecular Ecology 9: 1985–1996. Basidiomycetes on decomposing conifer trunks in northern Finland. Leake JR, Read DJ. 1997. Mycorrhizal fungi in terrestrial habitats. In: Karstenia 35: 1–51. Wicklow D, Söderström B, eds. The Mycota IV. Environmental and Såstad SM, Jenssen HB. 1993. Interpretation of regional differences in the microbial relationships. Berlin Heidelberg, Germany: Springer-Verlag, fungal biota as effects of atmospheric pollution. Mycological Research 97: 281–301. 1451–1458. Leavitt SW. 1993. Seasonal 13C/12C changes in tree rings: species and site Smith SE, Read DJ. 1997. Mycorrhizal symbiosis. London, UK: Academic coherence, and a possible drought influence. Canadian Journal of Forest Press. Research 23: 210–218. Söderström B, Read DJ. 1987. Respiratory activity of intact and excised Lilleskov EA, Hobbie EA, Fahey TJ. 2002. Ectomycorrhizal fungal taxa ectomycorrhizal mycelial systems growing in unsterile soil. Soil Biology differing in response to nitrogen deposition also differ in pure culture and Biochemistry 19: 231–236. organic nitrogen use and natural abundance of nitrogen isotopes. New Soponsathien S. 1998. Some characteristics of ammonia fungi 1. In relation Phytologist 154: 219–231. to their ligninolytic enzyme activities. Journal of General and Applied Lindahl B, Stenlid J, Olsson S, Finlay R. 1999. Translocation of P-32 Microbiology 44: 337–345. between interacting mycelia of a wood-decomposing fungus and Swift MJ, Heal OW, Anderson M. 1979. Decomposition in terrestrial ectomycorrhizal fungi in microcosm systems. New Phytologist 144: ecosystems. London, UK: Blackwell Scientific Publications. 183–193. Tanasaki E, Masuda H, Kinugawa K. 1993. Wood degrading ability of Michelsen A, Quarmby C, Sleep D, Jonasson S. 1998. Vascular plant N-15 basidiomycetes that are wood decomposers, litter decomposers, or natural abundance in heath and forest tundra ecosystems is closely mycorrhizal symbionts. Mycologia 85: 347–354. correlated with presence and type of mycorrhizal fungi in roots. Taylor AFS, Hogbom L, Hogberg M, Lyon AJE, Nasholm T, Hogberg P. Oecologia 115: 406–418. 1997. Natural N-15 abundance in fruit bodies of ectomycorrhizal fungi Nadelhoffer KJ, Fry B. 1988. Controls on natural 15N and 13C abundances from boreal forests. New Phytologist 136: 713–720. in forest soil organic matter. Soil Science Society of America Journal 52: Taylor AFS, Martin F, Read DJ. 2000. Fungal diversity in ectomycorrhizal 1633–1640. communities of Norway spruce (Picea abies [L.] Karst.) and beech Nadelhoffer KJ, Fry B. 1994. Nitrogen isotope studies in forest (Fagus sylvatica L.) along north–south transects in Europe. ecosystems. In: Lajtha K, Michener R, eds. Stable isotopes in ecology and In: Schulze E-D, ed. Carbon and nitrogen cycling in European forest environmental science. Boston, MA, USA: Blackwell Scientific ecosystems. Berlin Heidelberg, German: Springer Verlag, 343–365. Publications, 23–44. Tyler G. 1985. Macrofungal flora of Swedish beech forests related to soil Ohlsson AKE, Wallmark H. 1999. Novel calibration with correction for organic matter and acidity characteristics. Forest Ecology and Management drift and non-linear response for continuous flow isotope ratio mass 10: 13–29. spectrometry applied to the determination of δ15N, total nitrogen, δ13C Väre H, Ohenoja E, Ohtonen R. 1996. Macrofungi of oligotrophic Scots and total carbon in biological material. Analyst 124: 571–577. pine forests in northern Finland. Karstenia 36: 1–18. Pate J, Arthur D. 1998. δ13C analysis of phloem sap carbon: novel means Yamanaka T. 1999. Utilization of inorganic and organic nitrogen in of evaluating seasonal water stress and interpreting carbon isotope pure cultures by saprotrophic and ectomycorrhizal fungi producing signatures of foliage and trunk wood of Eucalyptus globulus. Oecologia sporophores on urea-treated forest floor. Mycological Research 103: 117: 301–311. 811–816.

© New Phytologist (2003) 159: 757–774 www.newphytologist.com 768 Research 27.5 0.2 27.627.5 0.4 0.3 26.0 0.3 25.0 0.2 25.4 0.4 25.3 0.5 26.6 0.4 24.5 0.3 26.9 0.6 25.2 0.3 26.424.8 0.7 0.1 − − − − − − − − − − − − − - northern Sweden; C 22.327.627.8–26.9 –28.6–26.7 – – 27.9–26.7 – 26.7–25.5 25.6–24.5 26.5–24.7 25.225.8–24.9 –27.6–25.0 – 25.5–24.0 27.8–24.6 25.325.8–24.6 25.6 –24.8 – –27.8 – – 28.7–24.7 24.9–24.7 – – – 13 Range Mean SE − − − − − − − − − − − − − − − − − − − − δ Picea. C% Range Mean SE Occasionally on 3

Salix. Occasionally on 2 N 15 Range Mean SE δ 5.83.1 –3.8–5.4 4.6 –4.3–4.8 –3.7–5.0 0.3 4.6 4.4 – 0.12.1–3.2 0.3 8.3 6.8–11.7 2.52.2–3.9 9.2 9.0 4.7–9.8 11.4–13.0 0.3 3.1 – 1.2 12.3 7.7 0.2 – 0.3 1.1 – 2.0–5.1 43.3–45.9 – 4.6–7.7 3.8 44.9 43.5–45.8 42.5–44.0 0.6 0.8 6.0 44.9 43.5 0.4 0.5 43.2 0.4 45.3–46.1 – 45.7 42.1–43.5 –3.6–5.1 0.2 4.3 42.9 – – 0.2 0.23.2 6.3–11.33.8 9.2 – 0.92.8–3.5 – 3.3 – 42.5–43.2 0.1 – 42.9 7.7 0.1 8.7–12.0 3.1 10.1 – 0.5 – – 41.5–43.6 – 42.1 41.6 0.3 41.7 – – – – N% Range Mean SE 1 4 4 4 4 6 5 1 1 6 n C and : in Sweden. (Åheden fungi collected from two forests of ectomycorrhizal (ECM) and saprotrophic N ratios in sporocarps 1 13 δ ?1 Betula Betula Betula Pinus Gen.Gen. 4 1 3.1–4.6 3.6Conifers 3.6 0.4 5Betula –Conifers 8.8–11.4 1 3.1–3.4 – 10.0Conifers 3.2 0.6 1 3.6 0.1 11.0 – 41.8–44.3 4.1 5.7–8.4Picea – 42.7 – 7.2 0.6 – – 0.5 – 3.1 41.0–41.6 43.0 12.3 41.3 – 0.1 – – – – – 42.1 42.9 – – – – Host N, %C, 15 ). δ

Fr. Conifers 2 3.9–3.9 3.9 – 9.9–11.1 10.5 0.3 (Fr.) Fr. (Fr.) Fr. Conifers 4 3.3–4.0 3.7 0.2 7.0–10.3 8.3 0.7 40.7–41.8 41.3 0.2

Watl. Fr. (L.) Hook Gen. 5 3.4–4.3 3.9 0.2 4.9–6.2 5.5 0.2 42.9–46.4 44.7 0.6 (Swartz ex Fr.) (Bon) Bon (Maire) Orton (Maire) Gen. 2 4.0–4.1 4.0 0.1 0.5–2.0 1.3 0.8 42.8–44.0 43.4 0.6 alues for %N, V richolomataceae Most probable host species; ?, mycorrhizal status doubtful, na, not applicable, Gen., generalist. Most probable Åheden Mycorrhizal species Boletales Boletaceae Chalciporus piperatus Bat. (Bull. ex Fr.) Leccinum molle Leccinum scabrum Gray S.F. (Bull. Ex Fr.) Leccinum variicolor Leccinum versipelle Snell (Fr.) Gomphidiaceae Betula Chroogomphus rutilus O.K. Miller ex Fr.) (Schff. Suillus variegatus O. Kuntze Agaricales Pinus T Tricholoma flavovirens (Pers. ex Fr.) Tricholoma virgatum Kumm. : Fr.) (Fr. Laccaria bicolor Amanitaceae Amanita muscaria Cortinariaceae Cortinarius armeniacus Fr. : Fr.) (Schaeff. Cortinarius armillatus Cortinarius bataillei (Moser) Høiland Cortinarius biformis Cortinarius camphoratus Fr. : Fr.) (Fr. Cortinarius evernuis Fr. : Fr.) (Fr. Cortinarius hemitrichus Fr. (Pers. : Fr.) Cortinarius laniger Cortinarius malachius Picea Betula Appendix 1 1 Stadsskogen, central Sweden

www.newphytologist.com © New Phytologist (2003) 159: 757–774 Research 769 25.225.8 0.4 0.2 24.9 0.3 23.8 0.09 22.0 0.6 24.6 0.2 27.3 0.5 26.4 0.8 27.1 0.8 27.6 0.4 24.9 0.4 − − − − − − − − − − − 24.8 − C 25.5 26.3–25.3 25.4–23.6 25.024.1 – .– –23.9–23.7 – 23.122.4 – – –24.0 – – – 22.6–21.4 25.4–24.0 24.728.2–26.4 29.2 –28.0–25.4 – 26.1 –27.9–26.3 .– – – 27.0 – – 28.8–26.5 25.225.6–24.2 – .– 13 − − − − − − − − − − − − − − − − − − − − − δ Range Mean SE Picea. C% Range Mean SE Occasionally on 3

Salix. 2.0 0.14 40.1–41.2 40.7 0.5 − 14.210.2 11.9 8.1 2.2 0.9 42.4–42.6 40.6–42.8 42.5 41.5 0.1 0.5 − − Occasionally on 2 N 1.9–2.2 1.7 – – 44.7 – – 15 − − Range Mean SE δ 2.7 – – 5.0–6.4 5.76.9 0.74.4–4.8 4.5 –3.4 9.2–9.42.9–5.6 0.1 9.3 4.0 – –4.73.7–4.6 0.1 0.4 7.5–11.4 4.1 – 9.0 – 17.5 0.4 2.6–9.5 42.4–42.8 1.2 – 42.6 – 7.3 2.2 0.2 9.2–14.2 42.2–42.9 1.1 11.7 – 5.3 42.5 – 2.5 0.2 42.1–45.9 – – 40.5 44.0 40.8–44.0 0.6 42.4 – 42.9 – 1.6 – 42.3 – – – – 3.1–3.3 3.3 0.1 9.7 3.5–4.9 4.0 0.2 4.2–6.7 5.7 0.4 42.2–43.5 42.7 0.2 N% Range Mean SE 1 1 3 1 6 1 2 6 n )2 1 Pinus Host nana 2 1 2.7–2.9 2.8na 5.0 0.1 – 1 –?2 2.0Pinus 1.4 –Betula – –Betula –Betula 0.2 42.3 – – – – 42.5 – – Betula Conifers 3 3.3–5.0 4.2 0.5 6.0–8.2 7.4 0.7 42.1–44.0 43.1 0.6 Mos. Conifers 1 4.6 – – 8.6 – – 42.9 – –

(L. : Fr.) Gen. 1 5.1 – – 14.1 – – 44.7 – – Fr. ( (Fr.) Snell (Fr.) Watl. (Fr. : Fr.) Fr : Fr.) (Fr. Gen. 1 3.7 – – 5.4 – – 43.7 – – Pil. & Dermek (Rostk.) Watl Britz Conifers 1 3.6 – – 6.3 – – 42.5 – – (Pers. ex Fr.) Gen. 4 3.0–4.8 3.9 0.4 5.9 (Scop.) Fr. Gen. 6 3.2–4.0 3.6 0.2 2.7–5.0 4.0 0.4 43.3–45.2 44.2 0.3 Bull. ex Fr. Gen. 5 4.4–8.0 6.5 0.6 6.1–10.3 8.5 0.9 39.7–43.1 41.7 0.7 richolomataceae Most probable host species; ?, mycorrhizal status doubtful, na, not applicable, Gen., generalist. Most probable Rozites caperata Karst. Russulales Russulaceae Lactarius rufus Russula paludosa Thelephorales Bankeraceae Phellodon tomentosa Baker species Saprotrophic Agaricales T Armillaria borealis Marx. & K. Korh. Strophariaceae Stropharia hornemanii Lund. : Fr.) (Fr. Cortinariaceae Gymnopilus penetrans Murr (Fr.). Coriolales CoriolaceaeTrametes hirsuta Pil. : Fr.) (Wulfen Stadsskogen Mycorrhizal species Boletales Boletaceae Chalciporus piperatus Bat. (Bull. ex Fr.) na Boletus edulis Boletus pinophilus Leccinum aurantiacum Gray (Bull. ex St. Am.) S.F. Leccinum holopus Leccinum scabrum Gray S.F. (Bull. ex Fr.) Populus Leccinum variicolor Leccinum versipelle Betula Cortinarius traganus (Fr. : Fr.) Fr. : Fr.) (Fr. Cortinarius scaurus Cortinarius semisanguineus Gill. (Fr.) Cortinarius strobilaceus Cortinarius pholideus 1

© New Phytologist (2003) 159: 757–774 www.newphytologist.com 770 Research 25.025.3 0.3 0.4 25.3 0.3 26.4 0.4 24.925.2 0.3 24.6 0.2 0.2 24.3 0.3 26.6 0.2 25.0 0.2 25.0 0.2 24.6 0.6 26.0 0.5 25.3 0.6 26.1 0.5 25.8 0.4 26.6 0.3 − − − − − − − − − − − − − − − − − C 25.8–23.8 26.2–24.4 25.926.1–24.4 –27.1–25.9 – 25.5–23.8 25.8–24.5 24.9–23.9 25.8–23.2 26.9–26.2 25.2–24.8 25.7–24.3 28.126.4–23.5 –26.7–24.1 – 26.3–24.4 26.7–25.6 24.326.3–25.4 – – 27.3–26.0 13 Range Mean SE δ − − − − − − − − − − − − − − − − − − − − Picea. C% Range Mean SE Occasionally on 3

Salix. 1.7 0.6 40.9–42.8 42.2 0.4 − Occasionally on 2 N 0.45–6.6 4.2 1.6 43.2–47.2 45.3 0.6 3.6–0.1 15 Range Mean SE − δ − .4–4.0 3.3.4–5.2.0–4.5 0.5 4.0.0–4.0 3.7 0.3 3.4.5–4.2 0.2 4.2–7.0 0.2 3.4 4.4–9.8 6.0 10.3–15.0 0.2 0.9 12.5 8.6–13.7 6.7 11.8 0.7 0.9 2.4–6.9 0.8 42.7–43.8 4.8 43.3 43.6–44.6 41.8–43.5 0.7 0.3 42.7–43.6 44.0 42.6 43.1 0.1 0.3 42.7–44.1 0.1 43.1 0.2 2.6–5.9 4.4 0.6 10.4–18.9 12.8 1.5 40.0–44.0 42.0 0.7 2.3–2.9 2.6 0.1 1.9–4.3 3.1 1.2 7.7–11.2 9.5 1.8 42.7–45.5 44.5 1.4 N% Range Mean SE 3.9 –2.3–4.1 3.2 – 0.3 6.43.5 – – – – 43.0 13.1 – – – – 41.1 – – 3.2–4.7 3.9 0.3 10.5–13.8 12.5 0.7 42.5–43.5 43.0 0.2 5 6 32 53 63 63 72 2 n 1 5 1 5 1 Pinus Gen.Gen. 4Pinus 1 4.5–6.0 5.2 4.5Pinus 0.4Pinus –Pinus 2.9–8.5Pinus – 5.6 1.2 6.7Pinus 43.0–43.5 – 43.2 0.1 – 42.8 – – Gen. 5 3.3–4.7 4.0 0.3 1.8–3.3 3.0 0.5 41.8–46.0 43.7 0.7 Conifers 5 3.2–5.4 4.6Gen. 0.4 2Conifers 0.6–5.5 2 3.6–4.0 3.1 3.8 1.0 3.3–3.4 0.2 3.3 41.4–42.5 0.1 7.2–8.2 41.9 0.2 2.7 7.8–8.6 0.5 8.2 0.4 44.0–44.4 44.2 42.1–42.5 0.2 42.3 0.3 Betula Host et al.

Watl.

(Batsch) Fr. Gen. 3 4.3–4.6 4.1 0.4 3.9–19.4 9.1 5.1 42.2–42.6 42.4 0.1

(Scop. ex Fr.) Gen. 3 4.5–6.0 5.1 0.5 2.4–3.5 1.6 2.0 42.3–43.0 42.6 0.2 (L.) Kuntze continued

L. : Fr. richolomataceae Most probable host species; ?, mycorrhizal status doubtful, na, not applicable, Gen., generalist. Most probable Appendix 1 Leccinum vulpinum 1 Xerocomus badius Gilb. : Fr.) (Fr. Xerocomus subtomentosus (L : Fr) Quél. Gomphidiaceae Chroogomphus rutilus O.K. Miller ex Fr.) (Schff. Gomphidius glutinosus Fr (Schff.) G. roseus Suillus bovinus Suillus luteus Picea Gray S.F. (L. ex Fr.) Suillus variegatus O. Kuntze (Swartz ex Fr.) Paxillaceae Paxillus involutus Rhizopogonaceae Rhizopogon obtextus S. Rauschert (Spreng) Strobilomycetaceae Tylopilus felleus Hygrophorus olivaceoalbus Picea (Bull. : Fr.) Karst. (Bull. : Fr.) Agaricales Hygrophoraceae Hygrophorus agathosmus Fr. (Fr.) Hygrophorus camarophyllus Dumée : Fr.) (Alb. & Schw. Picea Fr. ex Fr.) (Fr. T Laccaria laccata Bk. & Br. Tricholoma flavovirens (Pers. ex Fr.) Tricholoma fracticum(Britz.) Kreisel Tricholoma fucatum Kummer (Fr.) Tricholoma fulvum Pinus (DC : Fr.) Sacc (DC : Fr.)

www.newphytologist.com © New Phytologist (2003) 159: 757–774 Research 771 25.628.0 1 25.1 0.8 0.4 27.1 1.0 25.5 0.5 24.3 0.1 26.326.3 0.5 24.8 0.3 0.3 25.3 0.2 25.1 0.3 27.527.0 0.5 0.5 26.1 0.2 25.625.9 0.3 25.9 1.1 1.8 25.526.1 0.8 25.7 0.5 0.5 25.2 0.1 24.2 0.6 − − − − − − − − − − − − − − − − − − − − − − C 26.6–24.6 29.5–27.1 27.3–23.9 30.2–25.9 26 1–24.6 24.5–24.2 28.3–25.0 27.3–25.7 26.2–23.9 25.6–24.4 26.2–23.7 28.425.1 –28.4–26.7 27.9–26.4 – – 26.3–26.0 – 23.729.2 –26.0–24.8 27.0–24.9 – – 27.7–24.1 – 26.9–24.1 26.8–25.3 27.4–23.6 26.325.6–24.8 –25.8 – 25.3–23.1 – – 13 − − − − − δ Range Mean SE − − − − − − − − − − − − − − − − − − − − − − − Picea. C% Range Mean SE Occasionally on 3

Salix. Occasionally on 2 N 15 Range Mean SE δ 3.3–3.9 3.6 0.3 8.3–9.7 9.0 0.7 42.8–43.0 42.9 0.1 3.6–4.8 4.1 0.2 0.2–3.2 2.13.7 0.4 –4.0–6.2 41.2–46.14.2–5.2 5.2 – 44.1 4.6 0.6 0.8 0.3 10.7 8.4–10.5 8.6–10.12.9–3.9 9.1 –3.6–3.8 8.8 3.6 0.7 3.7 0.7 – 0.2 0.1 42.0–42.92.7–3.9 42.5–43.0 3.5 4.1–6.1 42.4 41.6 5.2–11.5 42.8 0.3 4.7 0.4 5.3 0.1 8.3 0.4 – 3.2 – 2.0–7.2 – – 42.4–43.2 5.0 41.7–41.9 42.8 1.6 41.8 0.2 0.1 5.4 42.5–44.1 43.2 – 0.5 – 42.4 – – N% Range2.7 Mean SE – – 9.4 – – 43.1 – – –26.9 – – 1 3 3 1 n 1 )6 )3 )2 )4 )2 1 Pinus Betula Betula Picea Picea Host Conifers 7Gen. 3.5–5.5 4 4.4 0.3 2.8–3.5 3.2 3.5–8.1 0.2 6.2 1.6–2.8 0.6 2.1 0.2 41.9–42.7 42.2 0.1 42.0–43.0 42.7 0.2 Conifers 8Gen.Gen. 3.1–5.2 5 4.0 7 0.3 3.3–5.1 4.2 3.9–5.8Betula 0.3 4.9 1.2–2.8( 0.3 2.1 3.5–5.2Conifers 0.2 2.2–5.5 4.3 1Gen. 0.3 44.4–47.0 4.1 45.6 4.0 1 0.6Conifers 0.3 40.4–44.3 – 2 42.1 2.7 41.6–45.7Picea 0.6 Conifers 43.4 – – 2.7–5.0 0.5 7 3.8 –Conifers 1.1 7.6 3.3–4.1 7Conifers 3.7 4.4–8.8 4.3 – 1 0.1 3.4–4.7 6.6 – 3.9 – 4.6 2.2 2.0–4.9 0.2 – 3.5 – 43.0 42.7–43.8 0.7 3.4–8.8 43.3 – 0.6 43.5 6.4 – 41.6–42.2 0.7 41.9 – 6.0 – 0.1 41.6–43.3 – – 42.8 0.2 – 45.1 – – Conifers 1 3.7 – – 5.4 – – 42.4 – – Gen. 2 3.7–5.0 4.4 0.7 9.8–11.6 10.7 0.9 42.9–43.4 43.2 0.3 Mos. Conifers 3 3.1–4.1 3.4 0.3 4.0–4.3 4.2 0.1 40.3–42.0 41.3 0.5 Peck . Fr. Conifers 3 3.9–4.3 4.1 0.1 5.8–7.1 6.3 0.4 40.6–42.1 41.4 0.4 Berk. Salix 3 3.4–5.9 5.0 0.9 1.7–6.2 3.2 1.5 42.4–42.7 42.6 0.1 (Fr.) Fr (Fr.) Fr. Conifers 3 3.3–3.5 3.4 0.1 5.6–8.8 6.4 1.0 41.7–42.6 42.4 0.4 Fr. (Conifers) 2 3.3–3.8 3.5 0.2 4.9–6.6 5.8 0.9 41.9–43.2 42.6 0.7 Fr. ( (Fr.) Fr. (Fr.) ( Fr. (

(L.) Hook Gen. 4 4.1–4.8 4.4 0.2 7.1–8.8 7.9 0.4 43.2–47.9 45.6 1.0 (Sch : Fr) Fr. ( Most probable host species; ?, mycorrhizal status doubtful, na, not applicable, Gen., generalist. Most probable Cortinarius uliginosus Cortinarius vibratilis Fr. : Fr.) (Fr. Hebeloma crustuliniforme Quél. (Bull : Fr.) Cortinarius traganus (Fr. : Fr.) Kumm. : Fr.) (Fr. Amanitaceae Amanita fulva Amanita muscaria Amanita porphyria (Alb & Schw : Fr) Mlady Amanita rubescens SF Gray (Pers : Fr.) Amanita virosa Bertillon (Kamarck) Cortinariaceae Cortinarius albo-violaceus Fr. (Pers : Fr.) Cortinarius armeniacus Fr : Fr.) (Schaeff. Betula Cortinarius armillatus Cortinarius bolaris Fr. (Pers : Fr.) Cortinarius brunneus Cortinarius camphoratus Fr. : Fr.) (Fr. Cortinarius crocea Big. & Guill. (Schff.) Cortinarius gentilis Cortinarius laniger Cortinarius limonius Fr. ex Fr.) (Fr. Cortinarius malachius Cortinarius muscigenus Cortinarius paleaceus Fr. (Weinm.) Cortinarius pholideus : Fr.)Fr. (Fr. Cortinarius semisanguineus Gill. (Fr.) Cortinarius speciosissimus Betula Kühner & Rom. Cortinarius stillatitius Cortinarius strobilaceus Tricholoma vaccinum(Pers. : Fr) Kumm. Tricholoma virgatum Picea 1

© New Phytologist (2003) 159: 757–774 www.newphytologist.com 772 Research 26.8 0.3 25.0 0.3 24.7 0.3 26.2 0.1 26.8 0.4 24.926.6 0.2 0.3 26.026.2 0.2 25.5 0.3 0.3 24.927.0 0.3 23.8 0.3 0.2 26.0 0.5 27.1 0.4 25.327.3 0.3 0.4 27.1 0.5 25.8 0.8 − − − − − − − − − − − − − − − − − − − C 27.2–26.5 26.226.527.025.4–24.5 – –26.5 – – – 26.9 – 25.7–24 0 – – – 26.2 – 27.7–25.7 25.6–24.0 27.9–25.3 26.2–25.8 26.8–25.3 26.3–24.3 25.1 – – 25.2–24.7 27.6–26.3 24.2–23.5 25.124.927.0–24.5 –26.4–27.5 – – 26.5 – 25.8–24.9 27.7–26.9 –28.1–26.6 – 24.2–26.7 23.5 – – 13 Range Mean SE δ − − − − − − − − − − − − − − − − − − − − − − − − − − − − − Picea. C% Range Mean SE Occasionally on 3

Salix. Occasionally on 2 N 0.2–3.00.3–4.5 1.70.7 1.4 0.6 0.8 – 43.9–45.1 – 44.1–45.2 44.5 44.7 0.2 0.2 45.4 – – 15 Range Mean SE δ − − − 4.0–4.4 4.2 0.23.6–4.23.4–4.9 3.9 4.2 7.6–9.4 0.1 0.2 8.5 0.9 3.8–9.5 6.4–8.9 6.5 7.5 44.2–44.6 0.9 44.4 0.3 0.2 44.4–46.8 43.9–46.8 45.8 45.2 0.4 0.3 N% Range Mean SE 3.2–5.02.2–4.9 4.1 3.93.3 0.9 0.5 – 2.6–4.5 3.5 – 0.9 43.8–46.5 45.2 1.3 3.5–3.63.2–5.0 3.6 3.8 0.13.1 0.4 6.2–6.6 – 2.5–5.8 6.4 4.3 – 0.2 0.73.2–3.4 3.3 1.6 44.8–44.9 43.9–45.6 44.9 0.1 44.53.5–4.0 0.1 – 3.8 0.4 7.1–8.1 0.2 – 7.6 0.5 4.3–6.3 43.6 5.5 45.0–45.2 0.6 – 45.1 0.1 – 43.6–44.8 44.2 0.3 2 6 8 n 2 5 1 2 4 1 2 3 2 2 1 Gen. 2 2.7–3.5 3.1 0.4 8.4–8.5 8.4 0.1 41.7–41.8 41.7 0.1 Gen.Gen. 3Gen. 1 4.6–5.1 5.0 5 5.3 0.2 – 3.9–4.6 4.3 4.2–4.3pinus –Picea 0.1 4.2 0.03 1.5 5.1–8.4 42.4–43.6 6.6 43.0 – 0.6 0.4 – 41.3–42.9 42.1 43.2 0.3 – – Host Betula Betula Betula Betula Pinus Gen.Alnus 5Gen. 3 3.2–3.9Picea 1 3.5 5.6–5.9Gen. 5.7 0.1 3.2Betula 0.1 3 0.5–4.2 –Betula 2.5 5.0–6.8 4.3–5.6 – 6.0 0.7 4.8 0.6 0.4 6.4 43.6–45.8 44.5 42.5–43.2 2.0–3.0 0.4 – 43.0 2.5 0.3 0.3 – 43.0–45.2 44.1 44.4 0.7 – –

Fr. Fr. Groeger Fr. Gen. 1 3.8 – – 5.4 – – 44.1 – – Quél. Conifers 5 3.0–3.9 3.5 0.2 (Fr.) Quél. (Fr.) Gen. 1 4.8 – – 3.5 – – 43.0 – – Hora Fr. Fr. Gen. 1 3.9 – – 6.6 – – 45.4 – – Fr. Fr. Gen. 4 3.2–5.1 4.3 0.4 5.9–8.2 6.9 0.6 42.5–45.4 43.7 0.7 Fr. Heim Conifers 1 4.8 – – 1.3 – – 43.4 – – (Scop.) Fr. Gen. 3 3.4–4.4 3.9 0.3 1.7–2.1 2.0 0.1 42.8–44.7 44.0 0.6 Heim. Conifers 1 3.8 – – 4.1 – – 42.9 – – Boud. Gen. 1 4.8 – – 2.9 – – 43.9 – – continued (Schff. ex Fr.) (Schff. camphoratus Gen. 5 4.0–5.1 4.5 0.2 1.9–5.4 3.8 0.6 43.9–45.3 44.8 0.3 Most probable host species; ?, mycorrhizal status doubtful, na, not applicable, Gen., generalist. Most probable Hebeloma mesophaeum 1 (Pers) Quél. Inocybe acuta Inocybe cincinnata Inocybe friesii Inocybe geophylla Kummer ex Fr.) (Sow. Inocybe pseudodestricta Stangl. & Veselsky Inocybe tigrina Rozites caperata Karst. (Pers. ex Fr.) Russulales Russulaceae Lactarius badiosanguineaKuehn. & Rom. Lactarius (Bull.) ex Fr. Picea Lactarius deliciosus Fr. Lactarius deterrimus Lactarius fuliginosus Lactarius vietus Russula atrorubens Russula betularum Lactarius glyciosmus Lactarius helvus Lactarius mitissimus Lactarius musteus Lactarius necator Karst. (Bull. em Pers. ex Fr.) Lactarius obscuratus (Lasch) Fr. Lactarius repraesentaneus Britz Lactarius rufus Lactarius scrobiculatus Fr. (Scop. ex Fr.) Lactarius theiogalus SF Gray (Bull : Fr.) torminosus Gray S.F. Lactarius trivialis Lactarius uvidus Appendix 1

www.newphytologist.com © New Phytologist (2003) 159: 757–774 Research 773 24.125.4 0.2 24.9 0.1 0.1 25.2 0.3 24.5 0.2 26.325.3 1.2 0.3 25.9 0.6 25.425.4 0.3 0.6 26.325.0 0.2 0.1 25.2 0.3 22.5 0.4 21.6 0.8 − − − − − − − − − − − − − − − C 24.4–23.9 25.7–24.9 24.9 24.827.726.4 –25.4 –25.4–24.9 – – 25.5 – 27.5 – – 26.024.9–24.1 – – –27.5–25.0 – – 25.6–25.1 – 26.3 – –26.5–25.4 – 25.8–24.9 27.2–23–2 26.6–26.0 24.9–25.0 25.9–24.4 22.9–22.2 23.322.7 – – – – 22.3–20.8 23.0 – – 13 δ Range Mean SE − − − − − − − − − − − − − − − − − − − − − − − − − − Picea. C% Range Mean SE Occasionally on 3

Salix. 0.03 0.2 42.4–43.3 42.9 0.4 3.0 0.2 41.0–42.2 41.6 0.6 − − Occasionally on 2 N 0.3–2.1 0.8 0.4 43.0–43.8 43.3 0.1 0.2–0.2 3.2–2.8 0.4 – – 43.3 – – 15 Range Mean SE − − − δ − 2.4 –3.7 – –3.44.0 – – 8.4 –2.9–3.8 3.3 – 4.4– – 0.4 – 7.9 – 0.5 4.1–4.7 – 4.4 – 45.6 – 0.3 – – 42.6 –2.8–3.0 45.4–47.4 2.9 46.42.7–3.7 46.3 – – 0.1 3.4 45.7 1.0 0.2 – – 2.3–2.8 – – 0.3–3.0 2.6 – 0.3 1.5 0.7 42.9–44.0 43.5 43.2–45.2 0.6 44.3 0.5 N% Range3.0–3.6 Mean 3.3 SE 3.6–4.1 0.3 3.8 0.2 5.0–5.1 1.5–5.4 5.1 0.1 3.43.6–4.0 2.0 3.8 41.9–42.7 42.3 0.1 43.63.2 0.4 7.1–10.5 – 43.6 9.0 0.1 – 1.0 43.9–44.8 44.4 0.3 1 1 1 2 2 n 2 2 3 1 )1 )4 2 3 3 1 Betula Picea Picea Pinus Host Betula Conifers 1Picea 3.7Picea –Pinus – 2.6 – –Picea and 43.3Conifers – 2Conifers – 1Gen. 4.6–5.6 5.1 5.8 1 0.5 –na 4.2 9.8–11.5na – 10.7 – 2 0.9 – 1 10.4 4.4–4.7 46.5–47.1 4.6 2.8 46.8 – 8.1 0.1 0.3 – – – – – 45.9 0.2 – 46.1 – – – – – 44.6 – – Pinus Fr. (

Fr. Fr. Gen. 3 2.9–3.2 3.0 0.1 5.1–5.5 5.3 0.1 44.0–44.2 44.1 0.1 L. : Fr. Gen. 3 2.4–3.1 2.7 0.2 7.0–9.6 8.6 0.8 43.1–43.9 43.5 0.2 Zv. Fr. Conifers 5 3.2–4.0 3.7 0.2 Fr. Gen. 6 2.6–4.1 3.2 0.2 7.9–9.8 8.9 0.3 43.4–46.0 44.6 0.4 (Fr.) Conifers 2 2.6–3.9 3.2 0.7 5.1–7.1 6.1 1.0 43.8–45.1 44.5 0.6 Schaeff. Britz. Conifers 2 3.3–3.8 3.5 0.3 2.6–5.0 3.8 1.2 43.2–43.4 43.3 0.1 Fr. ex Rom. Fr. Fr. Fr. Gen. 1 4.2 – – 1.8 – – 42.3 – – Fr.

Fr. Fr. Gen. 1 4.1 – – 8.4 – – 45.3 – – L. ex Fr. ( Lindbl. Conifers 2 3.2–4.1 3.7 0.4 richolomataceae Most probable host species; ?, mycorrhizal status doubtful, na, not applicable, Gen., generalist. Most probable Russula decolorans Russula emetica Russula foetens Russula gracillima Russula griseascens (Bon & Gaugué) L. Marti Russula integra Russula paludosa Russula puellaris Russula queletii Russula rhodopoda Russula sanguinea(Bull. ex St. am.) Fr. Russula sardonia Russula vinosa Russula xerampelina Fr. ex Secr.) (Schff. Cantharellales Pinus Albatrellaceae Albatrellus ovinus Kotl. & Pouz. Pinus Hydnaceae Hydnum repandum Hydnum rufescens Cantharellaceae Cantharellus cibarius Cantharellus lutescens Cantharellus tubaeformis Thelephorales Thelephoraceae Hydnellum ferrugineum Karst. : Fr.) (Fr. Hydnellum peckii Banker apud Peck Phellodon niger Karst. : Fr.) (Fr. Stadsskogen species Saprotrophic Boletales Paxillaceae Hygrophoropsis aurantiaca Mre. : Fr.) (Wulf. Paxillus atromentarius Fr. (Batsch : Fr.) Agaricales T Russula coerulea 1

© New Phytologist (2003) 159: 757–774 www.newphytologist.com 774 Research 25.6 0.6 22.9 0.1 23.5 0.2 22.0 0.3 24.0 2.2 22.2 0.5 − − − − − − C 24.4–26.4 22.9 – – 23.3 – – 23.0–22.9 23.9–23.1 20.823.3–22.7 23.8 – – 26.1–21.8 – – 22.8–21.7 24.5 – – 13 Range Mean SE δ − − − − − − − − − − − Picea. C% Range Mean SE Occasionally on 3

Salix. 2.1 0.5 41.6–43.0 42.3 3.0 1.6 42.2–43.2 42.7 0.5 − − Occasionally on 2 N 1.1–2.9 3.0 – – 44.1 – – 4.6–1.4 2.0 – – 44.1 – – 15 Range Mean SE − δ − − − N% Range Mean SE n 1 na 3 4.3–5.2 4.7 0.3 na 2 8.2–9.1 8.6 0.5 5.4–7.9 6.7 1.2 39.5–40.2 39.8 0.4 na 1 5.5 – – 1.0 – – 42.2 – – Host Bonord. na 1 9.0 – – (Fr. : Fr.) (Fr. na 2 3.8–4.5 4.1 0.3 2.5–6.0 4.3 1.7 42.8–43.7 43.3 0.5 S.F.Gray na 2 1.5–1.9 1.7 0.2 (Fr.) Sing. (Fr.) na 1 6.0 – – 2.0 – – 40.6 – –

Schaeff. na 2 8.1–8.5 8.3 0.2 4.9–6.8 5.9 0.9 39.2–40.8 40.0 0.8

Quél.0.2 na 3 7.3–8.9 8.0 0.5 1.5–2.7 2.2 0.4 35.9–40.4 38.8 1.5 (Fr.) Kummer (Fr.) na 1 5.1 – – (Pers. ex Fr.) na 1 6.6 – – 0.8 – – 43.7 – – continued ycoperdales ycoperdaceae Most probable host species; ?, mycorrhizal status doubtful, na, not applicable, Gen., generalist. Most probable Clitocybe clavipes 1 (Pers. ex Fr.) Kummer (Pers. ex Fr.) Mycena pura Kummer Mycena rosella Entolomataceae Clitopilus prunulus (Scop. ex Fr.) Kummer (Scop. ex Fr.) Entoloma nitidum Rhodocybe nitellina Agaricaceae Agaricus silvaticus Cystoderma carcharias Fay. (Pers. ex Secr.) Cortinariaceae Gymnopilus junonius Orton Hericiales Auriscalpiaceae Auriscalpium vulgare L L Lycoperdon foetidum Appendix 1