J. Jpn. Bot. 91: 205–217 (2016)

Ecological Patterns of Wood-Inhabiting Myxomycetes in a Natural Forest of the Kamikochi, the Hida Mountain Range, Central Japan

a, b Kazunari Takahashi * and Yuichi Harakon

aOkayama University of Science High School, 1-1, Ridai-cho, Kita-ku, Okayama, 700-0005 JAPAN; bHiroshima Municipal Hiroshima Secondary School, 1-14-1, Miirihigashi, Asakita-ku, Hiroshima, 731-0212 JAPAN *Corresponding author: [email protected]

(Accepted on February 11, 2016)

The ecology of lignicolous myxomycetes on fallen dead wood has never been clearly demonstrated in an intact forest. We compared the occurrence of myxomycete sporocarps on the dead wood of deciduous broad-leaved trees (BW) and coniferous trees (CW) to determine preferences for tree types and wood decay stages. Surveys of sporocarps were performed during summer and autumn 2011–2013 in a conserved forest of the Northern Japan Alps. We recorded 89 taxa (87 species, two varieties): 60 species on BW and 64 species on CW. Between seasons, the percentage similarity of myxomycete occurrence was low (0.239), while between tree types it was higher (0.463). Ordination of the assemblages using non-metric multidimensional scaling demonstrated that myxomycete occurrence changed seasonally and exhibited tree type preference. Forty- nine species (55% of the total) exhibited seasonal and/or tree type preferences. A majority of the species occurred on the intermediate decay stage of both tree types, but 38 species were correlated with particular decay stages of either CW or BW. The tree types contain wood parts at various decay stages, creating microenvironments that seasonally provide heterogeneous microhabitats for myxomycetes. Such diverse microhabitats on a local scale can promote myxomycete diversity in a natural forest.

Key words: Decay stages, myxomycetes, natural forest, preference, seasonality, tree types.

The myxomycetes (also known as the bodies large enough to be detectable in the field. true slime molds) are a group of eukaryotic Our understanding of myxomycete macro- microorganisms, usually appearing as a distribution and ecology comes mainly from fungus-like fruiting body that measures less studies performed in forests of temperate and than 1–2 mm in size (Martin and Alexopoulos neotropical terrestrial ecosystems (e.g., Schnittler 1969). They inhabit microenvironments in and Stephenson 2000, Takahashi and Hada forest ecosystems, such as the bark surface of 2010). Forest vegetation and climatic conditions living trees, forest floor litter, and fallen trees influence the distribution and diversity of (Stephenson 1988, 1989). The best-known myxomycetes (e.g., Rojas and Stephenson 2007, species are those that occur on dead wood, as Takahashi and Harakon 2010, 2012). A macro- these myxomycetes tend to produce fruiting ecological study showed that forest structure

—205— 206 植物研究雑誌 第 91 巻 第 4 号 2016 年 8 月 plays an important role in the structure of forest. myxomycete assemblages in neotropical regions (Rojas et al. 2011). Recent evidence related to Materials and Methods the distribution of microorganisms suggests Study site similarities to the distribution of higher plants The Kamikochi region is the southern and animals (Foissner 2006). However, the gateway to the Hida mountain range, southwest relationships between myxomycete distribution of the Nagano Prefecture in central Japan, and and particular ecological features have seldom remains undisturbed by anthropogenic activity. been studied on a local scale. The region is thus an excellent location for Dead wood from fallen trees represents studying the biodiversity of myxomycetes. It is a primary substrate in forest ecosystems and situated at around 1500–1600 m above sea level, constitutes a distinct microhabitat or micro- and shaped like a slender basin of around 1.5 km ecosystem for myxomycetes. Microhabitat in width and 10 km in length, and surrounded refers to a small, specialized habitat within a by mountains 2600–3000 m above sea level larger habitat (Stephenson 1989, Schnittler and (Fig. 1A). The region is snow-covered from Stephenson 2002, Kilgore et al. 2009). In one mid-November to early May, with temperatures study, microhabitats were more important than falling below −20 °C. During summer, daytime geographical locations or country in affecting temperatures reach around 22 °C (in August) myxomycete species distributions (Schnittler but can fall below 10 °C (36°15.2ʹN 137°40.1ʹE, and Stephenson 2000). Myxomycetes distribute 1529 m above sea level, http://ims.shinshu-u. varied microhabitats on fallen dead trees ac.jp/METIMS_WEB/wmz_weather_report. including differing tree types (Takahashi 2004, do#graph). Takahashi et al. 2009), decay states (Takahashi The vegetation of the survey area includes and Hada 2008b, Takahashi 2010), and wood deciduous broad-leaved forest and coniferous moisture content (Takahashi and Hada 2009). forest. The riparian deciduous forest site mainly Their distribution also change seasonally contains willows and larch species, including (Takahashi and Hada 2008a, Takahashi 2012) Populus maximowiczii A. Henry, Toisusu and over the course of several years (Takahashi urbaniana (Seemen) Kimura, and Salix rorida 2010). However, current knowledge remains Laksh. Approximately 50 years after the pioneer limited regarding how these distribution patterns trees, some betulaceous species also began are correlated with microhabitats in a natural growing, and these include Ulmus davidiana intact forest, because such forests are rare and Planch. var. japonica (Rehder) Nakai and Abies such studies have seldom been carried out. homolepis Siebold & Zucc. (Nagaoka and Therefore, reliable distribution data Okuda 2000). The coniferous forest site covers are needed from forests that have not been the lower part of the mountain slopes (exceeding disturbed by anthropogenic modifications. 1600 m alt.) and includes Abies homolepis, A. Information about myxomycete diversity and veitchii Lindl., and Picea jezoensis (Siebold & ecology in a natural forest promotes an in-depth Zucc.) Carriére var. hondoensis (Mayr) Rehder understanding of myxomycete distribution (Shin et al. 1999). At each survey site, we in a terrestrial ecosystem. The objective of observed around a hundred dead trees or logs on the present study is to analyze myxomycete the forest floor (Fig. 1B). species distribution patterns on the dead wood of deciduous broad-leaved trees and coniferous Sampling of myxomycete fruiting bodies trees, and to discern the relationship between the Intensive field surveys were conducted five species and the wood decay stage in an intact times in the summer (late July to mid-August) August 2016 The Journal of Japanese Botany Vol. 91 No. 4 207

Fig. 1. Study site in the Kamikochi valley where myxomycete fruiting bodies were surveyed. A. Landscape of Kamikochi valley in late May. B. A fallen log in the coniferous forest. C. Colony of favoginea fruiting bodies that occurred on deciduous decaying wood. D. Large colony of Tubifera casparyi fruiting bodies that occurred on coniferous decaying wood. and six times in the autumn (October) of 2011– More than 100 colonies of fruiting bodies 2013 (Table 1). These months coincide with were recorded on parts of dead wood for each the end of the two rainy seasons in Japan, when tree type. When the colonies of the same species the annual occurrence of myxomycetes peaks appeared within 30 cm of each other on a log, (Takahashi and Hada 2008a). We examined the areas were considered to represent a single myxomycete fructifications on dead wood (Fig. sample of scattered fruiting bodies that had 1C, D) from fallen trees on the forest floor. The developed from the same plasmodium (Eliasson survey was mainly carried out in the Myojin 1981, Stephenson 1988). Over 100 samples were area (1529 m), characterized by riverside forest observed for each tree type. Not all the fallen and deciduous broad-leaved trees (BW), as well trees that we examined contained a myxomycete as in the forest of Shimomatashirodani (1580– colony, but several colonies tended to occur on 1600 m), characterized by mountainous terrain a single log. The observed dead trees varied in and coniferous trees (CW). The two forest sites diameter (10–110 cm) and length (1–15 m). are located within 2.3 km of each other along Almost all of the fruiting bodies occurred on the drainage basin of the Azusa River in the decorticated parts of trunks, and several were Kamikochi. found on mosses growing on the trunks. 208 植物研究雑誌 第 91 巻 第 4 号 2016 年 8 月

Table 1. Sample number and species richness of microscopic examination were preserved in the myxomycetes during each survey period, showing herbarium of the Fukui Botanical Garden. species exactitudes associated with season and tree types in Kamikochi Analysis of myxomycete assemblages Species richness Number of We evaluated myxomycete assemblages in Survey Exactitude samples Sobs Sest (Sobs/Sest) each survey by comparing observed species richness (Sobs) to predicted species richness Summer 680 56 73 0.77 (Sest). The simplest nonparametric estimator B12S 120 26 29 0.91 B13S 110 20 27 0.74 Chao1 (Chao 1984) proposes precise estimation C11S 192 23 27 0.86 of predicted species richness from the observed C12S 110 17 20 0.84 species composition and richness (Joaquin et C13S 148 24 47 0.51 al 2006). Chao1 calculates the species richness of a sample using the equation Sest = Sobs + Autumn 827 63 75 0.84 a2/2b, where Sobs is the number of species in the B11A 115 19 26 0.73 sample, a is the number of singletons (species B12A 155 23 29 0.79 with only a single occurrence in the sample), B13A 109 16 37 0.43 and b is the number of doubletons (species with C11A 146 18 21 0.86 exactly two occurrences in the sample). The C12A 126 21 28 0.74 C13A 176 25 30 0.83 relative exactitude of the Sobs was evaluated by comparing Sobs with Sest using the following Total 1507 89 99 0.90 equation: index of exactitude = Sobs/Sest. BW* 609 60 65 0.92 Myxomycete assemblages associated CW* 898 64 77 0.83 with each season and with each tree type Italics indicate sum of seasonal and whole survey. were compared by calculating the percent *BW. Deciduous broad-leaf wood. CW. Coniferous wood. similarity of two assemblages. The equation for percentage similarity, which considers relative Species identification species abundance, is PS = Σ min (a, b, c,…x), Specimens of common and easily recognized where min = the lesser of the two percentage myxomycetes were usually identified in the field compositions of species a, b,…x in the two via their sporocarp structure, observable under assemblages (Stephenson 1988). a magnifying glass. When species were difficult Ordination of the myxomycete to recognize in the field, part of the colony assemblages was carried out using non-metric was sampled and preserved in a paper box. multidimensional scaling (NMDS; Kenkel and Detailed identification was then performed in Orlóci 1986), with appropriately analyzed and the laboratory. The morphological characteristics ranked myxomycete assemblages (Takahashi of the fruiting bodies were examined under a 2010). We used PAST (Hammer et al. 2001; stereomicroscope and other traits useful for http://folk.uio.no/ohammer/past/) to compute species identification were further examined the ordination for Bray-Curtis similarity (Bray under a microscope. Identification followed and Curtis 1957), which measures the similarity the nomenclature of Yamamoto (1998, 2006) between sample sets. and more recent information from Lado (2005– We performed Fisher’s exact probability test 2015). The frequency of individual species of independence on data to compare species occurrence was recorded and collected in an frequency on BW versus CW, or across seasons. on-site assemblage during each survey. The When the occurrence of a species on any type voucher specimens that had been identified with of tree or in any season was significantly higher August 2016 The Journal of Japanese Botany Vol. 91 No. 4 209 than zero (p < 0.05), the species was regarded 35 species were represented by four or fewer as frequent and was considered to prefer that samples (Table 2, p. 253). Species richness was specific type of tree or season. Observations greater in autumn than in summer, and species based on fewer than five records were excluded were relatively more diverse on CW than on from the analysis. BW. Seven species were considered dominant on dead wood in the study forest because they Comparison of wood decay stages constituted at least 3% (45) of the total sample In the field, we noted the hardness of the number: Trichia decipiens (Pers.) T. Macbr., wood part from which myxomycete specimens (L.) Fr., Physarum were collected, following previously described viride (Bull.) Lister, Stemonitis axifera (Bull.) methods (Takahashi and Hada 2009, Takahashi T. Macbr., Metatrichia floriformis (Schwein.) 2010). Wood hardness, which indicates the stage Nann.-Bremek., M. vesparium (Batsch) Nann.- of physical decay, was measured using a soil Bremek., and Tubifera ferruginosa (Batsch) hardness tester (Fujiwara Scientific Company). J. F. Gmel. The dominant species for BW The experimenter measured the resistance value were L. epidendrum, M. floriformis and M. of a spring pressed into the wood (8.0 kg of vesparium, while the dominant species for CW sustained force/40 mm depth in the wood). The were P. viride, S. axifera, Colloderma oculatum pressure (kg/cm2) needed to drive the spring X (C. Lippert) G. Lister, and Lamproderma mm into the wood was calculated according to columbinum (Pers.) Rostaf. the formula p = 100X / [0.7952(40 – X)2]. To compare the mean hardness for the number of Seasonality and preference for tree types myxomycete species, we performed one-way The 11 myxomycete assemblages from analysis of variance followed by Tukey’s HSD our survey were grouped based on the results test (p < 0.05), using Excel version 5.0 (Esumi of the NMDS analysis. Values for two axes of Co. Ltd., 2001). Species recorded with four or the NMDS showed the ordinated positions of fewer samples were excluded from the analysis. the assemblages on two-dimensional factors, relating the first axis to season and the second Results axis to tree type (Fig. 2). The myxomycete Species richness and dominant species assemblages were regressed against a distinct A total of 87 species and two varieties hierarchic ranking of four groups. The first two representing 25 genera were identified from groups comprised autumnal assemblages that 1507 records (hereafter, any mention of species discriminated between BW and CW, whereas the number includes varieties); 56 and 63 species other two groups were summer assemblages that were obtained in summer and autumn, while also differed in terms of tree type. Thus, while 64 and 60 species were found on CW and BW, season explained assemblage similarity more respectively. Mean indices of exactitude (Sobs/ effectively than tree type, assemblage make-up Sest) were 0.77 in summer, 0.84 in autumn, 0.92 shifted depending on season and preferred tree for BW, and 0.83 for CW (Table 1). These data type. The percentage similarity of assemblages allowed us to estimate that 99 species exist on between the seasons was low at 0.239, and the dead wood in the intact forest of the Kamikochi, similarity value between the two tree types was indicating that our survey missed some species. 0.463, with greater similarity in summer (0.459) Species were arranged based on the amount than in autumn (0.418). of significant samples for each season (Table Species replacement occurred in the 2). Of these, 54 species were represented by myxomycete assemblages depending on the five or more samples (Table 2, p. 252), whereas season (Table 2). Only 30 of the 89 species were 210 植物研究雑誌 第 91 巻 第 4 号 2016 年 8 月

Table 2. Myxomycete species and number of samples arranged for each season and tree type in the Kamikochi. Seasons included summer and autumn, and tree types were deciduous broad-leaved wood (BW) and coniferous wood (CW). Species are listed according to their statistically significant abundance on each season and across tree types a Wood typesb Species Orderc Season Total Summer Autumn BW CW Species recorded with five or more samples Physarum viride (Bull.) Lister P 98** 1 22 77** 99 Lycogala epidendrum (L.) Fr. L 93* 79 82* 90 172 Stemonitis axifera (Bull.) T. Macbr. S 57** 1 7 51** 58 cancellata (Batsch) Nann.-Bremek. L 40** 1 6 35** 41 Physarum flavicomumBerk. P 39** 3 36** 39 Tubifera ferruginosa (Batsch) J. F. Gmel. L 38** 7 21 24 45 Ceratiomyxa fruticulosa (O. F. Müll.) T. Macbr. C 27** 16* 11 27 Physarum nutans Pers. P 18** 5 19** 4 23 Hemitrichia calyculata (Speg.) M. L. Farr T 18** 4 19** 3 22 Lycogala exiguum Morgan L 18** 9 9 18 Stemonitopsis gracilis (G. Lister) Nann.-Bremek. S 16** 5 11 16 Stemonitopsis typhina var. similis (G. Lister) Nann.-Bremek. & Y. Yamam. S 14** 3 3 14 17 Trichia lutescens (Lister) Lister T 14** 1 15** 15 cinerea Fr. T 13** 1 6 8 14 Comatricha nigra (Pers. ex J. F. Gmel.) J. Schröt. S 12** 12** 12 Cribraria tenella Schrad. L 11** 10** 1 11 Stemonitopsis hyperopta (Meyl.) Nann.-Bremek. S 10** 2 8 10 Fuligo septica (L.) F. H. Wigg. P 10* 4 2 12* 14 Cribraria intricata Schrad. L 8** 8** 8 Cribraria intricata var. dictydioides (Cooke & Balf. f.) Lister L 5* 5 5 Cribraria languescens Rex L 5* 5 5 Trichia decipiens (Pers.) T. Macbr. T 43 225** 90 178* 268 Metatrichia floriformis (Schwein.) Nann.-Bremek. T 3 51** 51** 3 54 Metatrichia vesparium (Batsch) Nann.-Bremek. T 1 46** 46** 1 47 Colloderma oculatum (C. Lippert) G. Lister S 43** 43** 43 Lamproderma columbinum (Pers.) Rostaf. S 2 37** 39** 39 Reticularia splendens (Morgan) T. Macbr. L 1 29** 18* 12 30 Trichia erecta Rex T 29** 2 27** 29 Physarum flavidum (Peck) Peck P 27** 24** 3 27 Trichia favoginea (Batsch) Pers. T 4 26** 21** 9 30 Diderma ochraceum Hoffm. P 21** 21** 21 Tubifera casparyi (Rostaf.) T. Macbr. L 4 20** 24** 24 Cribraria macrocarpa Schrad. L 1 16** 3 14 17 Hemitrichia clavata (Pers.) Rostaf. T 11** 11** 11 Cribraria purpurea Schrad. L 10** 1 9* 10 Diderma deplanatum Fr. P 7* 2 5 7 Diderma umbilicatum Pers. P 7* 6* 1 7 Cribraria meylanii Brândza L 6* 6* 6 Diderma radiatum (L.) Morgan P 6* 6* 6 Lepidoderma tigrinum (Schrad.) Rostaf. P 6* 6* 6 Diderma lucidum Berk. & Broome P 5* 5 5 Elaeomyxa cerifera (G. Lister) Hagelst. P 5* 5 5 Fuligo candida Pers. P 5* 2 3 5 Arcyria denudata (L.) Wettst. T 5* 4 1 5 Calomyxa metallica (Berk.) Nieuwl. T 5* 5 5 Brefeldia maxima (Fr.) Rostaf. S 5* 5** 5 Trichia scabra Rostaf. T 3 5 8** 8 Stemonitis fusca Roth S 5 1 6** 6 Trichia subfusca Rex T 6 2 8* 8 Reticularia lycoperdon Bull. L 2 9 6 5 11 Lamproderma arcyrionema Rostaf. S 2 6 3 5 8 Cribraria vulgaris Schrad. L 1 5 1 5 6 Stemonitopsis typhina (F. H. Wigg.) Nann.-Bremek. S 3 2 3 2 5 Cribraria rufa (Roth) Rostaf. L 1 4 5 5 aSignificantly different (*p < 0.05, **p < 0.01) between Summer and Autumn. bSignificantly different (*p < 0.05, **p < 0.01) between deciduous broad-leaved wood and coniferous wood. cC. Ceratiomyxales. E. Echinosteriales. L. Liceales. P. Physarales. T. . S. Stemonitales. August 2016 The Journal of Japanese Botany Vol. 91 No. 4 211

Table 2. Continued a b c Season Wood types Species Order Total Summer Autumn BW CW Species recorded with four or less samples Lycogala conicum Pers. L 4 4 4 Comatricha pulchella (C. Bab.) Rostaf. S 4 4 4 Didymium minus (Lister) Morgan P 3 3 3 Didymium squamulosum (Alb. & Schwein.) Fr. & Palmquist P 3 3 3 Arcyria obvelata (Oeder) Onsberg T 3 3 3 Trichia persimilis P. Karst. T 2 2 2 Stemonitis pallida Wingate S 2 1 1 2 Cribraria splendens (Schrad.) Pers. L 2 2 2 Physarum penetrale Rex P 1 2 3 3 Stemonitis flavogenitaE. Jahn S 1 1 2 2 Physarum globuliferum (Bull.) Pers. P 1 1 2 2 Badhamia nitens Berk. P 1 1 1 Arcyria ferruginea Saut. T 1 1 1 Comatricha tenerrima (M. A. Curtis) G. Lister S 1 1 1 Cribraria filiformisNowotny & H. Neubert L 1 1 1 Cribraria piriformis Schrad. L 1 1 1 Physarum subnutans Y. Yamam. P 1 1 1 Trichia contorta (Ditmar) Rostaf. T 1 1 1 Enerthenema papillatum (Pers.) Rostaf. S 1 1 1 Trichia verrucosa Berk. T 3 3 3 Diderma chondrioderma (de Bary & Rostaf.) G. Lister P 3 2 1 3 Cribraria dictyospora G. W. Martin & Lovejoy L 3 3 3 Physarum newtonii T. Macbr. P 3 3 3 Badhamia macrocarpa (Ces.) Rostaf. P 2 2 2 Physarum nudum T. Macbr. P 2 2 2 Arcyria imperialis (G. Lister) Q. Wang & Yu Li T 2 2 2 Cribraria argillacea (Pers. ex J. F. Gmel.) Pers. L 2 2 2 Physarum pusillum (Berk. & M. A. Curtis) G. Lister P 2 2 2 Physarum didermoides (Pers.) Rostaf. P 1 1 1 Physarum leucopus Link P 1 1 1 Arcyria helvetica (Meyl.) H. Neubert, Nowotny & K. Baumann T 1 1 1 Hemitrichia serpula (Scorp.) Roataf. T 1 1 1 Lamproderma arcyrioides (Sommerf.) Rostaf. S 1 1 1 Barbeyella minutissima Meyl. E 1 1 1 Arcyria monticola Y. Yamam. & H. Hagiw. T 1 1 1 aSignificantly different (*p < 0.05, **p < 0.01) between Summer and Autumn. bSignificantly different (*p < 0.05, **p < 0.01) between deciduous broad-leaved wood and coniferous wood. cC. Ceratiomyxales. E. Echinosteriales. L. Liceales. P. Physarales. T. Trichiales. S. Stemonitales. common between summer and autumn. Of the differed depending on tree type (Table 3). 89 species, 46 (52%) occurred seasonally (p < Trichiales and Liceales orders were both 0.05), with 21 in the summer and 25 in autumn. diverse; from the former, 17 species were found In addition, when comparing between tree on BW and 13 on CW, while from the latter, type preferences across seasons, seven species 12 species were found on BW and 20 on CW. preferred BW and six species preferred CW Thus, myxomycete taxonomic order was clearly during summer, while eight species preferred related to particular tree types. BW and 10 preferred CW during autumn (Table 2, p. 252). Ultimately, 34 species (38% in Relationship with decay stage of wood total) significantly preferred a certain tree type, Myxomycete species were distributed including 17 species on BW and 17 species on on a spectrum of woods, ranging from non- CW (p < 0.05). decayed hard parts to decayed soft parts. Their Species composition of taxonomic orders occurrence pattern was a unimodal distribution 212 植物研究雑誌 第 91 巻 第 4 号 2016 年 8 月

Fig. 2. Arrangement of myxomycete assemblages using non-metric multidimensional scaling (NMDS) analysis based on the first two scores. A total of 11 assemblages were separated into four groups associated with seasonality and tree type. Abbreviations of assemblage names are shown in Table 1. Triangles indicate coniferous wood (CW) and circles indicate deciduous broad- leaved wood (BW). that peaked for moderate decay stages regardless Table 3. Comparison of species richness and mean wood of tree types (Fig. 3). The mean values of wood hardness (depth in mm) for species across five orders that occurred on deciduous broad-leaved wood (BW) hardness (with standard deviation) were 19.7 and coniferous wood (CW). The same letters indicate ± 6.7 mm for BW and 20.8 ± 7.2 mm for CW, no significant difference p( < 0.01) with no significant difference between them. The Number of Wood hardness majority of species preferred the intermediate Order species (mm depth) stage of wood decay. BW CW BW* CW* Species belonging to the same taxonomic Physarales 18 19 24.5 ± 5.3a 24.0 ± 7.1a group tended to occur on wood at a particular Stemonitales 12 11 19.1 ± 6.1bc 20.8 ± 7.2bc stage of decay (Table 3). They were broadly Liceales 12 20 18.9 ± 6.9bc 18.4 ± 7.1c separated into three stages of wood hardness bc c preference with the Tukey’s independent test. Ceratiomyxales 1 1 18.7 ± 7.7 18.1 ± 9.6 c b Physarales species occurred on hard wood Trichiales 17 13 18.7 ± 6.4 20.9 ± 6.2 of both wood types, followed, in order, by *a, b, bc, c: the same letter indicates no significant difference (p < 0.01). Stemonitales, Liceales, and Ceratiomyxales species. However, species in Trichiales from BW preferred softer wood than those from CW. Trichia lutescens (Lister) Lister and Physarum Thirty-eight species (based on four or more nutans Pers. occurred on harder wood, whereas samples each) were related with particular Cribraria intricata Schrad. and C. tenella wood decay stages depending on the tree type Schrad. were found on softer wood. Among CW (Table 4). For example, the BW specialists specialists, Trichia subfusca Rex, Comatricha August 2016 The Journal of Japanese Botany Vol. 91 No. 4 213

Fig. 3. Normal distribution of myxomycete colonies with relation to hardness of deciduous broad- leaved wood (BW) and coniferous wood (CW). nigra (Pers. ex J. F. Gmel.) J. Schröt., and is the first report to focus on the distribution Physarum flavicomum Berk. occurred on harder of wood-inhabiting myxomycetes in an intact wood, while Tubifera casparyi (Rostaf.) T. forest of Japan. Woody debris of fallen trees Macbr. Cribraria purpurea Schrad., and C. plays an ecological role maintaining biodiversity meylanii Brândza were found on softer decayed of microbes in forest ecosystems (Rayner and wood. Thus, we observed strong preferences for Bobby 1988). Our data on species richness in the a particular stage of decay within a specific tree Kamikochi forest (89 species in total, consisting type. of 60 species on BW and 64 species on CW) The other 12 species generally preferred a reached 90% exactitude, allowing for the similar hardness on both tree types (exceptions quantitative analysis of species distribution on were Stemonitopsis gracilis [G. Lister] Nann.- dead wood. In comparison to previous studies, Bremek and Cribraria cancellata [Batsch] we found comparable species richness for the Nann.-Bremek). For example, Physarum BW in the Kamikochi forest, species richness viride (Bull.) Lister and Reticularia splendens was about the same as the richness recorded in (Morgan) preferred harder wood, whereas the beech forest of Mt. Daisen, western Japan Tubifera ferruginosa (Batsch) J. F. Gmel., (Takahashi 2001), at 60 species versus 62 Lycogala exiguum Morgan, and Arcyria species, respectively. On the other hand, four cinerea Fr. preferred moderately decayed wood. species on CW were remarkably abundant in the Consequently, most of these 12 species inhabited Kamikochi (p < 0.01), compared with 52 species microenvironment niches primarily defined by in an adjacent subalpine coniferous forest of wood decay stage, but not tree type (Table 4). Mt. Yatsugatake, Central Japan (Takahashi and Harakon 2012), i.e., T. decipiens, Lycogala Discussion epidendrum, Physarum viride, and Colloderma Species diversity in an intact forest oculatum. The present study provides important Tree type (BW versus CW) is known to knowledge for understanding microbial influence myxomycete assemblages in summer biodiversity in forest ecosystems because it across four distant forests in Japan (Takahashi 214 植物研究雑誌 第 91 巻 第 4 号 2016 年 8 月

Table 4. Species preference for tree type on deciduous et al. 2009). Here, we found a lower percentage broad-leaved wood (BW) and coniferous wood (CW). similarity of 0.463 between assemblages on the Species are arranged in order of mean wood hardness two tree types than that of 0.63 in the summer (depth in mm) with standard error assemblages (Takahashi et al. 2009). Thus, we BW CW consider variation in tree types and seasonality Mean* SD Mean* SD within a single forest to affect myxomycete Specific species on BW species diversity. The ecological patterns that a Trichia lutescens 28.1 3.1 emerged from our analysis have improved our Physarum nutans 25.3ab 4.5 abc understanding of how myxomycetes use wood Physarum flavidum 22.8 5.6 parts of different tree types as microhabitats. Hemitrichia clavata 20.3abcd 4.9 Diderma umbilicatum 19.3bcd 3.9 Breferdia maxima 19.2cd 7.8 Seasonality Hemitrichia calyculata 18.6cd 5.1 Species assemblages differed between Stemonitis fusca 17.5cd 5.4 summer and autumn according to tree type in Metatrichia floriformis 17.0d 5.7 the Kamikochi forest (Fig. 2). Seasonal patterns Metatrichia vesparium 16.8d 6.2 in myxomycete occurrence have been verified Trichia scabra 16.4d 6.3 d in the forests of Japan (e.g., Takahashi 2001, Cribraria intricata 13.5 4.9 Takahashi and Hada 2008a). Our data confirm Cribraria tenella 12.5d 3.7 the previous reports of seasonality. Of note, Specific species on CW preference for BW or CW among species in the Trichia subfusca 29.6a 5.7 autumnal assemblage was quite noticeable, with Comatricha nigra 27.3ab 4.8 Physarum flavicomum 27.3ab 5.2 a percentage similarity of 0.418 between the Colloderma oculatum 23.3abc 5.5 assemblages on BW and CW, compared with Lamproderma columbinum 20.7bcd 8.0 0.459 during the summer. Fuligo seprica 20.1bcde 9.3 Seasonal change characterized the occurrence Diderma ochraceum 20.0cde 6.9 of several species on the different tree types. Cribraria macrocarpa 19.7cde 5.9 Tubifera casparyi showed a significant Lepidoderma tigrinum 18.7cde 6.9 cde preference for CW in late summer to autumn. Trichia erecta 18.5 6.1 Species in Trichiales also exhibited considerable Tubifera caspalryi 17.2de 7.4 de specificity (e.g., Hemitrichia calyculata and Cribraria purpurea 13.0 3.8 Cribraria meylanii 10.6e 3.4 Trichia lutescens were found on BW in summer, Other species and Metatrichia floriformis, M. vesparium, H. Physarum viride 26.2a 3.5 26.4ab 6.4 clavata, and T. favoginea were found on BW Reticularia splendens 26.1a 5.8 25.2abc 6.9 in autumn, whereas T. erecta was significantly Stemonitopsis gracilis 22.0abcd 6.8 15.0de 7.9 found on CW in autumn). Trichia decipiens abcd de Stemonitis axifera 20.6 4.8 17.5 6.6 was unusual in its specificity; this species Lycogala exgiuum 20.4abcd 5.1 21.0abc 9.6 abcd bc appeared in late summer on CW but proliferated Arcyria cinerea 20.4 9.8 19.5 5.7 with dominance on both tree types in autumn. Trichia decipiens 20.0bcd 5.3 21.3bc 5.5 Lycogala epidendrum 19.3cd 6.4 20.0cde 6.2 Still others exhibited spatially and temporally Ceratiomyxa fruticulosa 18.7cd 7.7 18.1cde 9.6 specific occurrence, such as Lamproderma Tubifera ferruginosa 17.0cd 5.3 17.3de 7.3 columbinum on the intermediate decay stage of Trichia favoginea 16.0d 6.9 17.2de 8.1 CW in autumn. Our study confirmed that certain Cribraria cancellata 10.0d 1.6 16.3de 6.6 species were seasonally distributed on particular *a, ab, abc, abcd, bc, bcd, bcde, cd, cde, d, de, e: the same tree types, suggesting that seasonal segregation letter indicates nosignificant difference p( < 0.01). increases myxomycete diversity in forests. August 2016 The Journal of Japanese Botany Vol. 91 No. 4 215

Distribution pattern on tree types and decay myxomycete choice of microhabitat. stages We observed differences in myxomycete Ecological patterns of myxomycete species distribution patterns on both tree types at the In total, 49 species (55%) were specialists level of the order: Liceales species strongly that exhibited specific ecological preferences preferred CW, while Trichiales preferred BW for season and/or tree type (Table 2). Our (Table 3). In addition, on BW, the five taxonomic data corroborate previous field observations groups preferred differing wood hardness that have also found species preferences for a on BW, as follows (in order of preference certain tree type (Takahashi et al. 2009) and for decreasing wood hardness): Physarales decay stage (Takahashi 2010). However, we > Stemonitales > Liceales > Ceratiomyxales do note some contradictions with previous = Trichiales (Table 3). In contrast, on CW, research, specifically with regard to Cribraria Liceales used well-decayed wood, and spp., reported to occur mostly on well-decayed Trichiales occurred on similar wood hardness as coniferous wood (Ing 1994). Here, we found Stemonitales. that Cribraria species were found not only on On the species level, Tubifera ferruginosa, softer wood but also on intermediate wood (C. Lycogala exiguum, and Arcyria cinerea showed macrocarpa) and hard wood (C. languescens). no apparent preferences for either tree type, In addition, several Cribraria species used well- whereas 34 species demonstrated a clear decayed wood depending on the season (e.g., in preference (Table 2). Among the species with a summer, C. cancellata was found on soft CW preference, 26 specifically preferred particular and C. intricata and C. tenella were found on decay stages on the different tree types (Table soft BW, whereas in autumn, C. purpurea and 4). From these patterns, we concluded that C. meylanii were found on softer CW). Of the the varied preference for different tree types 15 Cribraria species, six showed a significant and wood decay stages promote myxomycete preference for a particular microhabitat diversification in an intact forest. associated not only with wood properties, but It is unknown why myxomycetes prefer also with season. certain woody substrates, but the reason may These wood and seasonal preferences are involve the interaction of physical and biotic inconsistent with the hypothesis of cosmopolitan factors within the decaying wood. The wood distribution, proposed because myxomycete parenchyma in and angiosperms differs spores are small (around 10 μm in size for all not only in fundamental structure but also in species) and easily dispersed, principally by physicochemical composition (Scheffer and wind. Indeed, several species of myxomycetes Cowlig 1966, Rayner and Boddy 1988). The have a nearly global distribution (Stephenson size, number, and distribution of tracheids and et al. 2008); for example, Arcyria cinerea is vessels are important factors that determine present throughout most of North and South spatial patterns of decomposer mycelial America. However, our study found that invasion, which strongly influence cryptogam A. cinerea was distributed on both types of communities on dead wood (Fukasawa et al. wood but was limited to moderately decayed 2015) by potentially allowing the intrusion of wood in the summer. Moreover, reports on myxamoebae and plasmodia, the growing phase other species reveal that they are restricted to of the myxomycete life cycle. In conclusion, tree specific geographic areas (Estrada et al. 2013). type and the stage of wood decay are associated Together, our results and previous findings with particular wood physicochemical of restricted distribution support a recently properties that are highly important in affecting proposed hypothesis that free-living terrestrial 216 植物研究雑誌 第 91 巻 第 4 号 2016 年 8 月 protists exhibit local endemism (Foissner 2006). Hammer Ø., Harper D. A. T. and Ryan P. D. 2001. Therefore, the presence of diverse vegetation PAST: Paleontological Statistis Software Package for Education and Data Analysis. Palaeontologia within a single forest, including deciduous Electronica 4(1). http://palaeo-electronica.org/2001_1/ broadleaf trees and coniferous trees, creates past/issue1_01.htm. numerous microhabitats that can promote Ing B. 1994. Tansley Review No. 62: The phytosociology the myxomycetes diversity and abundance in of myxomycetes. New Phytol. 126: 175–201. temperate regions. Joaquin H., Paulo A. V. B. and Clara G. 2006. Evaluating the performance of species richness estimators: We conclude that the spatial and temporal sensitivity to sample grain size. J. Anim. Ecol. 75: usage patterns of myxomycetes are related to 274–287. microhabitat distribution, which emerge from Kenkel N. C. and Orlóci L. 1986. Applying metric and differences in vegetation on a local scale. To nonmetric multidimensional scaling to ecological studies: some new results. Ecology 67: 919–928. understand the ecological characteristics of Kilgore C. M., Keller H. W. and Ely J. S. 2009. Aerial myxomycetes in greater depth, we require reproductive structure of vascular plants as a studies on biogeographical patterns of their microhabitat for myxomycetes. Mycologia 101: 305– distribution, the restriction of species to certain 319. Lado C. 2005–2015. An on line nomenclatural information microhabitats, and potential antagonism with system of Eumycetozoa. http://eumycetozoa.com/data/ decomposers. index.php (Accessed [6 July 2015]). Martin G. W. and Alexopoulos C. J. 1969. The A grant for financial support from the Myxomycetes. University of Iowa Press, Iowa. Ministry of Education, Culture, Sports, Science Nagaoka F. and Okuda S. 2000. Species composition of and Technology for Environmental Science Abies homolepis forests in relation to landforms along the Azusa River around Kamikochi, Central Japan. (2011, subject number 24-1262-004), is Veg. Sci. 17: 55–72. gratefully acknowledged. We are thankful to Rayner A. D. M. and Boddy L. 1988. Fungal the staff of the Ministry of Environment and Decomposition of Wood: Its Biology and Ecology. the Agency for Cultural Affairs for giving us John Wiley & Sons, Chichester. Rojas C. and Stephenson S. L. 2007. Distribution and permission to survey the conservation area of the ecology of myxomycetes in the high-elevation oak Kamikochi. forests of Cerro Bellavista, Costa Rica. Mycologia 99: 534–543. References Rojas C., Stephenson S. L. and Huxel G. R. 2011. Bray J. R. and Curtis J. T. 1957. An ordination of upland Macroecology of high-elevation Myxomycete forest communities of southern Wisconsin. Ecol. assemblages in the northern Neotropics. Mycol. Prog. Monogr. 27: 325–349. 10: 423–427. Chao A. 1984. Nonparametric estimation of the number of Scheffer T. C. and Cowlig E. B. 1966. Natural resistance classes in a population. Scand. J. Stat. 11: 265–270. of wood to microbial deterioration. Annu. Rev. Eliasson U. 1981. Patterns of occurrence of myxomycetes Phytopathol. 4: 147–168. in a spruce forest in south Sweden. Holarctic Ecol. 4: Schnittler M. and Stephenson S. L. 2000. Myxomycete 20–31. biodiversity in four different forest types in Costa Rica. Estrada-Torres A., Wrigley de Basanta D. and Lado C. Mycologia 92: 626–637. 2013. Biogeographic patterns of the myxomycete Schnittler M. and Stephenson S. L. 2002. Inflorescences of biota of the Americas using a parsimony analysis of Neotropical herbs as a newly discovered microhabitat endemicity. Fungal Divers. 59: 159–177. for myxomycetes. Mycologia 94: 6–20. Foissner W. 2006. Biogeography and dispersal of Shin N., Ishikawa S. and Iwata S. 1999. The mosaic microorganisms: a review emphasizing protists. Acta structure of riparian forest and its formation pattern Protozool. 45: 111–136. along the Azusa River, Kamikochi, central Japan. Jpn. Fukasawa Y., Takahashi K., Arikawa T., Hattori T. and J. Ecol. 49: 71–81. Maekawa N. 2015. Fungal wood decomposer activities Stephenson S. L. 1988. Distribution and ecology of influence community structures of myxomycetes and Myxomycetes in temperate forests. I. Patterns of bryophytes on coarse woody debris. Fungal Ecol. 14: occurrence in the upland forests of southwestern 44–52. Virginia. Can. J. Bot. 66: 2187–2207. August 2016 The Journal of Japanese Botany Vol. 91 No. 4 217

Stephenson S. L. 1989. Distribution and ecology of habitat preference of myxomycetes in relation to the myxomycetes in temperate forests. II. Patterns of decay state of dead Fagus crenata wood from a natural occurrence on bark surface of living trees, leaf litter, beech forest in western Japan. Hikobia 15: 155–164. and dung. Mycologia 81: 608–621. Takahashi K. and Hada Y. 2009. Distribution of Stephenson S. L., Schnittler M. and Novozhilov Y. K. 2008. Myxomycetes on coarse woody debris of Pinus Myxomycete diversity and distribution from the fossil densiflora at different decay stages in secondary forests record to the present. Biodivers. Conserv. 17: 285–301. of western Japan. Mycoscience 50: 253–260. Takahashi K. 2001. Ecological distribution of myxomycetes Takahashi K. and Hada Y. 2010. Geographical distribution on beech forest on Mt. Daisen in the south-west of of myxomycetes on coniferous deadwood in relation to Japan. Bull. Kurashiki Mus. Nat. Hist. 16: 55–64 (in air temperature in Japan. Mycoscience 51: 281–290. Japanese). Takahashi K., Hada Y. and Mitchell D. W. 2009. Substrate Takahashi K. 2004. Distribution of myxomycetes on preference of lignicolous myxomycetes relative to different decay states of deciduous broadleaf and wood types in temperate Japanese forests. Hikobia 15: coniferous wood in a natural temperate forest in the 287–298. Southwest of Japan. Syst. Geogr. Plants 74: 133–142. Takahashi K. and Harakon Y. 2010. Ecology of wood- Takahashi K. 2010. Succession in myxomycete inhabiting myxomycetes in three different types of communities on dead Pinus densiflora wood in a natural forests on the Kirishima Mountains of south- secondary forest in southwestern Japan. Ecol. Res. 25: western Japan. Hikobia 15: 499–510. 995–1006. Takahashi K. and Harakon Y. 2012. Comparison of wood- Takahashi K. 2012. Myxomycete seasonal occurrence and inhabiting myxomycetes in subalpine and montane species diversity associated with different decay states coniferous forests in the Yatsugatake Mountains of of coarse woody debris. Hikobia 16: 131–140. Central Japan. J. Plant Res. 125: 327–337. Takahashi K. and Hada Y. 2008a. Seasonal change of species Yamamoto Y. 1998. The Myxomycete Biota of Japan. Toyo composition and dominant species of myxomycete Shorin Publishing Co., Tokyo (in Japanese). communities on dead wood of Pinus densiflora in Yamamoto Y. 2006. Supplement of “The myxomycete biota warm temperate forest. Hikobia 15: 145–154. of Japan.” The Japanese Society of Myxomycetology, Takahashi K. and Hada Y. 2008b. Occurrence patterns and Fukui Botanical Garden (in Japanese).

a b 高橋和成 ，原紺勇一 ：北アルプス上高地の自然林に おける腐朽木生変形菌の生態 多くの変形菌は倒木などの腐朽木上に発生するが，森 計量多次元尺度構成法により季節と樹種の違いで区分 林内での種多様性や生態についてはよく分かっていな された．全出現種の 55％にあたる 49 種が，季節性や樹 い．本研究の目的は，北アルプスの特別保護地区にある 種選好性を示した．多くの種は適度に腐朽した木質に出 上高地で，落葉広葉樹と針葉樹の腐朽材に出現する変形 現し，特定の腐朽段階に出現する種として 38 種が認め 菌を季節的に調査し，変形菌の季節性と樹種および木質 られた．変形菌は森林内の腐朽木に季節的に発生し，樹 腐朽段階への選好性を明らかにすることである．調査は 種や木質の腐朽段階を選び分けて分布している．こうし 2011–2013 年に行い，本地域に 87 種 2 変種の変形菌が た生態的分布パターンにより，森林生態系における変形 分布することが明らかになった．変形菌の種群の夏と秋 菌の種多様性が維持されていると考えられる． の類似性は，百分率類似度で 0.239 となり低い値であっ （a岡山理科大学附属高等学校， た．落葉広葉樹には 60 種，針葉樹には 64 種が出現し， b広島市立広島中等教育学校） 樹種間の類似度は 0.463 であった．変形菌の種群は，非