Fungal Diversity Fungal succession on bamboo in Hong Kong Dequn Zhou1* and Kevin D. Hyde2 IFaculty of Conservation Biology, Southwest ForestiyCollege, Kunming, Yunnan, P.R. China 650224 2Centre for Research in Fungal Diversity, Department of Ecology & Biodiversity, The University of Hong Kong, Pokfulam Road, Hong Kong, SAR Zhou, D.Q. and Hyde, K.D. (2002). Fungal succession on bamboo in Hong Kong. In:Fungal Succession (eds. K.D. Hyde and E.B.G. lones). Fungal Diversity 10:213-227. Fungal succession on Bambusa tuldoides has been studied in Hong Kong. Fungal communities changed over time during the decay process. Based on sporulation of fungi, the fungal communities on bamboo baits can be categorized into early colonisers, middle-stage colonisers, later colonisers, regular inhabitants and sporadic inhabitants. Fungal communities on naturally dead bamboo and baits comprised rare species and mainly middle-stage colonisers. Seasonality had an effect, as more fungi were present during the wet season. Rainfall positively impacted on fungal occurrence, but temperature and relative humidity appeared to have little influence. Anthostomella species are regular inhabitants of bamboo, being dominant throughout the observation period and probably play a dominant role in its decomposition. Key words: bamboo baits, colonisers, fungal community, seasonality. Introduction Studies of fungal succession on plant substrates are well documented (Wildman and Parkinson, 1979; Kuter, 1986; Frankland, 1992, 1998; Tokumasu et al., 1994), but fungal succession on bamboo is poorly documented. Leung (1998) placed several bamboo baits in terrestrial habitats in Hong Kong and observed a succession of fungi occurring on these baits. Sixty fungal taxa (including 33 unidentified fungi), comprising 16 ascomycetes and 44 anamorphic fungi were identified during a one year exposure period. The observation time, however, was relatively short and fungal succession was probably still at an initial stage after one year. In temperate regions fungal succession on bracken litter would probably not cease until after 5-6 years, when bacteria are dominant and 95% dry mass reduction is estimated to take 11-23 years (Frankland, 1976, 1998). In tropical regions, the time for complete decay of a plant may be relatively shorter than in temperate regions, as there is a greater fungal diversity and these fungi have more intensive activity (Stevens, *Corresponding author: Dequn Zhou; email: [email protected] 213 1989). The size of bamboo baits in Leung (1998), were also small (10 x 2 cm), so that the majority of fungi in the succession process may not have been sampled. A survey over a longer period using larger bamboo baits is therefore desirable. In this paper, succession and diversity of fungi on naturally dead bamboo samples and baits of Bambusa tuldoides at Tai Po Kau Nature Reserve, Hong Kong was investigated. It is understood that succession as observed here, is the sequential sporulation of fungi on a subsrate. The relationships between fungal succession and climatic factors were also analyzed. Materials and methods The study site covering about 0.3 hectares was selected in a bamboo forest at Tai Po Kau Nature Reserve, New Territories, Hong Kong, which comprised two bamboo species, i.e., Bambusa tuldoides and B. shiuyingiana. The bamboo forest lies on the bank of Tai Po Kau forest stream near the lookout post, at the field office of Taj Po Kau Nature Reserve, Department of Agriculture and Fishery, Hong Kong SAR. Bambusa tuldoides was selected for the fungal succession study, because this species occupies a major proportion of the forest, about 70% of the total coverage. On 23 August 1998, 114 healthy bamboo culms were randomly cut, labeled and left to rot naturally on the forest floor to be exposed as baits for fungal succession. Every two months, four whole bamboo culms were randomly retrieved and cut into samples ca. 25 cm long. Fifty of these samples were randomly selected for study at each period. At the same time, one standing healthy bamboo culm in the site was also . collected as a control. This was also cut into samples 25 cm long and 10 samples were randomly selected. Fifty naturally dead bamboo culm samples (ca. 25 cm long) were also arbitrarily collected from the same site at the same time, in order to compare the fungal diversity with that of the succession samples. A total of 1320 naturally dead bamboo, healthy bamboo samples and bamboo baits were collected during the two-year observation period. The samples were returned to the laboratory, where they were incubated in zip-lock polythene bags lined with moistened tissue. The samples were checked after three days, one week, one month and two months for fungal fruiting bodies. Squash mounts and sections of fungal fruiting bodies were mounted in water for observation and measurement. All bamboo samples were collected based on random sampling techniques using arandom table. Data analysis, i.e., diversity index (Shannon- Wiener index) (Shannon and Weaver, 1949), evenness index (Ludwig and Reynolds, 1988), species-area 214 Fungal Diversity curves, species richness and similarity indices (Bray-Curtis coefficient and S~rensen's index) (Bray and Curtis, 1957) were calculated using the equations. Equitability: it is most commonly expressed as Pielou's evenness index (1') (Ludwig and Reynolds, 1988): l' = H/Hmax Where H = Shannon- Wiener Index; Hmax = Log2S' l' = 1when the community structure is perfectly even (i.e. all taxa are found in l' an equal number of samples) and = 0 (or close to 0) when the community structure is at the extreme of uneven (i.e. only one taxon presents). S' = species richness Percentage frequency of occurrence = Number of samples on which agiven taxon occurred x100% number of samples examined Shannon-Wiener index (diversity index): This index was used to calculate diversity (Shannon and Weaver, 1949): Where H = the probability of finding each taxon in a collection, and Pi = the number of individuals in ith species. This index can also be calculated with the program developed with W.Q. Wang. Species similarity matrix: a matching triangular array of similarities between every pair of species, in terms of patterns of occurrence across the samples. The two most useful in ecology is Bray-Curtis coefficient (Bray and Curtis, 1957) and S~rensen' sindex. Bray-Curtis coefficient: the similarity between jth and kth samples was calculated as follows: _ 2:;=12minL·u'Y) -100· G Sk}-1001 ~2:i=1 \Yij + Yik) 2:i=p1 Iu}I+YY.,) ik " - (- 2:;=I\Yi) ..- YiJ J 215 Where = score (count or biomass) for ith species injth samples (i = 1,2,..,p; Yij j = 1,2,3, n). Species-area curve: This was also plotted to estimate the minimum number of bamboos that it was necessary to establish the community in any habitat. Species richness (S'): the number of species present in any given area. Results and discussion Fungal diversity Fifty-seven fungal taxa were found on the bamboo baits and this was higher than that on naturally dead bamboo samples (38 species). The average number of fungi occurring on each bamboo bait was 2.1, but on each naturally dead bamboo sample it was 0.8 (Table 3). The diversity index (Shannon Wiener index) for fungal communities on bamboo baits was 2.4, while for naturally dead bamboo this was 2. The similarity indices between the fungal communities occurring on naturally dead bamboo samples or bamboo baits during different periods were generally low (Tables 5, 6). Possible reasons for the lower numbers of fungi on naturally collected bamboo samples are discussed later. In general, the fungal diversity on bamboo appears to be lower than that on the other woody substrata in terrestrial or aquatic conditions (Gamundi et al., 1987; Aoki et al., 1990; Leung, 1998; Ho et al., 2002). The reasons for this are unclear. Anamorphic fungi (34 spp.) were the dominant group of fungi on the bamboo baits during the observation period, followed by ascomycetes (22 spp.) and 1 basidiomycete (Tables 2, 3). The fungal communities on naturally dead bamboo samples were almost equally composed of ascomycetes (20 spp.) and anamorphic fungi (18 spp.) (Tables 1, 3). Marasmius rotula (basidiomycete) appeared on the bamboo baits (as a sporadic inhabitant) after 18months. Previous authors have also reported that anamorphic fungi were dominant during succession studies. Leung (1998) found that anamorphic fungi (44 spp.) were dominant over ascomycetes (16 spp.) on the baits of Bambusa sp. in Hong Kong and he also identified an unidentified basidiomycete occurring on the bamboo baits during the later stage of the succession. Tsuneda (1983) suggested that the reason why basidiomycetes occurred late in fungal succession was because of their relatively low saprobic competitive abilities for the water- or solvent-soluble constituents of plant materials. Several authors have also reported that anamorphic fungi were dominant over other taxonomic groups in the succession process of other substrates. Tokumasu et al. (1994) 216 0025024000422420040227006000020682044000000022200040000000000041100000000000200400200104000004000000000204000060202000200000000122040440000000000000004000002000000000000400006000040480000000040002000002200202000022023620000000006000000000020000000004066286202000000002000000000282000220000000002000200002000003800600000000020260020202060000201121011110100002008010181111111120100024042100200001030000608826111.066..6.........2126040206......406.8.7660824242245.05428010464.57666642540665811347637488 WDWreeytt 0400020000000020000000400S000001000002020..e77ason1s0and collection dates Fungal Diversity AAAAAAAAAAAAAAACECDRPRPMPFHEESTEROGPEESVr·n..r.p..ss..rlaoofp.oou..ehov.lnueylthi:itttlatifemerihdalrrhecscaxudssuihoratedtpinapbsmscudraylooisatroroyoydrgnnsobosooesiionouaerselrpeeicsimdsrclnppniaennasmesp:enmlneuorsmgoclpceoiyhoxplttohlanroerlghoutniofiropiataollbsheaiyuliutiobyuriesriraanluirnmsssotnhihntmmleliaealasslemarucimdaaiapr-laloaaarailrastqesainaWjeisieselgohianturonarnuaidstrulsmimaggmceishslimeeiaeaoiimebbapnssceposlemusknnihnarlippaaoohrdep.nlamlrbsabcysciaiao.mmnlcai.ietruaaysnlalcecosrotieerabtmbbmiaamilipaeneomoroduugr.isfbusombpnbiissuainpauinfuosgaaaouttneeeiseiisisaeekbssiodlddraiiieitaclemcmeieseuorseoncxsmiupnulsaamlgirnesraaxence 16 10/98 12/98 03/99 05/99A0c7r/o9d9ic0ty9s/99ba1m1/b9u9si0co2l/a00 04/00 D06r/y00 09/00 Table 1.Frequency of occurrence of fungi on naturally dead bamboo. Species 217 Table04200420642002284320400000400066000000000000000000640000000000000602400040040000400040000000000040000000000000002000400000006742004244000000000002400002040000600210000000000000002080000008000002000000000202000060000600000800002000000500280000000000002020000000002002022000002000000022200000010000200000028000034222007002000082000000020000000008362322232061000063660562002088000200000000011111080815181110084101431101101111111111112480408600004041407289306704066606023066626438600422. FrequencyWDWreeyottf occurrence00000000000000000000000000S00000000000000000eafsuonngsi oanndbamcoblloeoctiboanitsd.ates AAAEEDAAAAADDAAAAAAMFEEEECDPPPPMMNMHGGGCOO.rr..pr.n..·..s.uu..lihleooiioiylrihf.xatto.alseedteeitilfitsthhsoaiereprrpbemypmsscohaomsclsurnrystsyoauadloorraimpitrhdsyosasoahosoescperoommneirraiilateeaserguhopaolmmnoocsolmpsnneeoaoeimleicpeoanluroermmisulotetoiiipmalhiihlmnpnuratbohibhymngsidilpduulasomyrnahnisaeaoleisisamtumiipacaeyioyloammsaouctiaasirlaaireutaresnouehetpssclrsserrpasesjtuimaitpiereamomeaieiitaalegsfeusemaaeipsplgmt!.rheioesbplaeaemhbsl,sessmutrirrnumllpbao.sarbpuaioilatcsszusiaa.umnrtspa.mmikltpetuaeeusmpeurtosumeaib.laabpdznmmp.dslnoarrueagbu.ipe.fpisinenaaysuanuisa.tutaitlueateesasiarsiensdaeilmeintiimieorsiisieafnpuulalisemenexnsussis 10/98 12/98 03/99 05/9A9c0ro7d/9ie9ty0s9/9b9am1b1u/9s9ico0l2a/00 04/00 D06r/y00 09/00 Species 218 0298423420010000000000000000424000000600644000642000004400000850020005321000042003802000002030020000000000003252600200080040231000008201011110011111111324466.4440825822200868404634806.62454624247 02..6662 0WDW2..5r7eey71tt 02..5998 03..600/00000000S00000026eas02o..n521s670a0n2.d.6552col02le..63c17tion01..49d83ate0s1..5928 Fungal Diversity RRTRESSSSSPSTVVoapp·...o.hv.e.fumioltryeurpmrnaiasorrnunoiauligsiaicnnlbdddpsdoeghcaeitifeeaieliruclreuslcsrlocoeaalsalmmeoaarqtatnrsiueniudipsumscbeeiomchunhorhumcytarrysmyyteeosurosobmiapmofauaimaodstloacbelrucesuiumsicricroeolnalace 0 10/98 12/98 03/99Sh0a5n/n9oP9no-0d7oW/s9pi9eonr0ei9ur/m9i9nde1el1xo/n9g9a0tu2m/00 04/00 D06r/y00 09/00 Species Table 2 continued. Table 3. Fungal species diversity on naturally dead bamboo and bamboo baits. Naturally dead bamboo Bamboo baits samples Ascomycetes 20 21 Basidiomycetes 0 1 Anamorphic fungi 18 35 Species richness 38 57 Species per sample 0.75 2.1 Shannon index 2.0 2.4 S0rensen's index 0.8 * *: This figure shows similarity between fungal communities on dead bamboo samples and bamboo baits. Table 4.Ecological classification of fungi on bamboo baits. Successional group Taxonomic group Total Ascomycetes Basidiomycetes Anamorphic fungi Early colonisers 2 0 6 8 Middle-stage colonisers 8 0 6 14 Late colonisers 1 0 2 3 Regular inhabitants 3 0 3 6 Sporadic inhabitants 7 1 18 26 Total 21 1 35 57 219 observed fungal succession on pine needles in Germany and identified 1 ascomycete, 1basidiomycete, 62 anamorphicfungi and 9 zygomycetes. Similar results have also been reported onNothofagus dombeyi leaf litter (Gamundi et al., 1987),on momi fir needles (Aokiet al., 1990)and submerged wood (Ho et al., 2002). Successional replacement Fungi replace one another at different stages of succession and this has been observed by several authors (Frankland, 1976, 1998; Leong et al., 1991; Leung, 1998;Tokumasu, 1998;Ho et al., 2002). Frankland (1976) found that weak parasites first dominated dead bracken, followed by primary saprobes, then secondary saprobes and finally common soil fungi. Ho et al. (2002) categorized the fungi that occurred on submerged wood baits into 3 groups, i.e., pioneer, early and later successional groups. Leung (1998) divided fungi identified on bamboo baits into two groups, i.e., early colonisers and regular inhabitants. The temporal replacement of fungi in the present study can be categorized into (1) early colonisers, i.e., those fungi occurring during the first 4 months and disappearing thereafter; (2) middle-stage colonisers, i.e., those dominant during 6-10 months and disappearing thereafter; (3) later colonisers, which became dominant after 10 months and persisted until the end of the experiment; (4) regular inhabitants, which occurred throughout or mostly throughout the experimental period, and (5) sporadic inhabitants, which occurred on bamboo baits sporadically (Table 4). In all of these studies, succession has been correlated with identification of sporulating fungi and this may not reflect what takes place within the bamboo. It would be extremely difficult, however, to prove that succession is taking place rather than observation ofa sequential appearanceoffruitingbodies. One procedure would be to cut up the bamboo samples, surface sterilize them, place them on agar and isolate fungi growing from the samples. Such a method however, would have many problems, asthe sampleswould need to be cut into very small units andmost ofthe isolates wouldprobablybe sterilemycelia. In general, most fungi on bamboo baits were rare species, which were encountered once or twice. The dominant species mostly occurred on bamboo baits from ea. 6 months onwards, while quite a few species were present throughout the observation period (Table 2). The dominant early colonisers were Apiospora sinensis, Arthrinium luzulae, A. phaeospermum, coelomycete sp.,Fusarium sp.,Petroeonium sp.,Roussoella pustulans and Veronaea indiea; while Acrodietys bambusieola, A. ereeta, Anthostomella suleigena, Apiospora montagnei, Astrosphaeriella australiensis, Cordella johnstonii, Eutypella gliricidiae, Gilmaniella bambusae, Monoehaetia karstenii, Oxydothis 220 Fungal Diversity Table 5. Similarity analysis (Bray-Curtis coefficient) of fungal communities on naturally dead bamboo samples. Date 10/98 12/98 03/99 05/99 07/99 09/99 11/99 02/00 04/00 06/00 09/00 10/98 100 46 11 16 16 12 10 40 17 14 20 12/98 100 13 9 14 12 16 36 24 12 15 03/99 100 14 5 17 14 30 11 23 19 05/99 100 23 51 17 13 12 47 38 07/99 100 22 18 28 39 38 43 09/99 -tOO 40 15 29 62 47 11/99 100 16 33 29 32 02/00 100 27 27 23 04/00 100 49 51 06/00 100 57 09/00 100 Table 6. Similarity analysis (Bray-Curtis coefficient) of fungal communities on bamboo baits. Date 472474924444040004818819434500032132136455133015705221701247103980723420212038967960503804/98* 1100230//9989 0150/909 10117/90/9909 00219/00/9009 041/0000 061/0000 09/10000 100 100 100 *000000001195732469211//////////99990000O999999000089/98's value is zero (no fung1i0/9o8ccurred when the healthy sampling bamboos were nceuwt)l.y oraniopsis, Ramichloridium musae, RoussoiWa hysterioides, Spirodecospora bambusicola and Thyridium chrysomallum were middle-stage colonisers. Later colonisers were Astrosphaeriella stellata, Ellisembia bambusae and E. bambusicola; while Anthostomella contaminans, A. flagellariae, Podosporium elongatum, Astrosphaeriella jissuristoma, Phaeoisaria clematidis and Podosporium nilgirense were regular inhabitants (Table 2). There were 26 sporadic inhabitants, which occurred on bamboo baits (Tables 1, 2). These groups of fungi occurring during different periods are summarized in Table 4. Marasmius rotula (a basidiomycete) was identified during the later stages of succession (Tables 2,4). Frankland (1998) has reviewed previous studies and explained the possible mechanisms for the successional of fungi on substrata. She thought there were several possible causes: availability of space and species of 221 differential performance; processes and phenomena conditioning, the first e.g. dispersal, competition between species and animal grazing involving whole communities: inherent characters that defined the outcome of the latter and operated between populations and individual mycelia, e.g. differential rates of growth and nutrient uptake. These explanations may partially account for successional occurrence of fungi. The most compelling activities during fungal succession may be with inherent characters operating between populations and mycelia among different fungal species through differential rates of growth and nutrient uptake. Unfortunately this is impossible to observe in situ. Where do the early colonisers come from? The issue of where the early colonisers on various substrata come from has yet to be clearly established (Frankland, 1992, 1998, Leung, 1998; Norden et al., 1999). The early colonisers may be endophytes, as endophytes exist asymptomatically in all tissues of living plants (Bacon and White, 2000). Umali et al. (1999) studied endophytes in Bambusa tuldoides at Tai Po Kau, Hong Kong. They identified 23 species and 14 "morphospecies" of mycelia sterilia. Of the eight early colonisers found on bamboo baits in this study (Table 2), Arthrinium phaeospermum, Fusarium sp. Phomopsis sp. and Podosporium elongatum were also found as endophytes (Umali et al., 1999). Apiospora sinensis, Arthrinium phaeospermum, coelomycete sp., Fusarium sp., Roussoella pustulans and Veronaea indica, found on the bamboo baits, were also found on naturally dead bamboo samples. Fusarium spp. and Mucor spp. were also found on baits in this study, after they were incubated in zip-lock polythene bags. They are also possibly endophytes. The similarity indices (S0rensen's index) between fungal communities on bamboo baits and naturally dead bamboo samples were high (0.84) (Table 3). The two fungal communities were very similar, with many fungal species common to both substrates (Tables 1, 2). Thus the early colonisers may therefore arrive from other nearby substrata, such as naturally decaying bamboos. Endophytes, however, can not be excluded, as four species identified by Umali et al. (1999) were also found as early colonisers on bamboo baits and many of the endophytes isolated in traditional studies cannot be identified and remain as mycelia sterilia (Umali et al., 1999). Comparison offungal diversity on bamboo baits and naturally dead bamboo Species richness and frequency of occurrence of fungi on naturally dead bamboo samples were lower than that on bamboo baits (Tables 1-3). It is likely that the bamboo sampled in this study died several years ago. According to the field workers at Tai Po Kau Nature Reserve, many bamboo plants were blown over in 1996 by a typhoon. Thus the bamboo plants at the site may have been 222