Nature and Human Society: The Quest for a Sustainable World (1997)
Chapter: Estimating the Extent of Fungal Diversity in the Tropics
Page 156
Estimating the Extent of Fungal Diversity in the Tropics.
Introduction
With the rapid global destruction of tropical habitats, many peopleincluding conservationists, research scientists, and those wishing to use biodiversityare beginning to recognize that we should find out what we are destroying before it is too late. Tropical deforestation has become the crucible of today's extinction crisis (Wildman 1997), but we should not forget that many other habitats are under threat.
But why in particular do we need to measure fungal diversity? Why do we even want to know which fungi are present in an ecosystem? Why not just measure their isozyme activity or use molecular techniques to indicate fungal presence? Mycologists have robust answers to such probing questions (Hawksworth 1991, 1993, 1998; Hyde 1996a,b; Lodge and others 1996), but conservationists and ecologists, let alone the broader public and politicians, are rarely appropriately briefed. Fungi are important in biological control, in medicine, in biotechnology, in bioactive novel compounds, in decomposition, in nutrient cycling, as actual and potential food resources, in enzyme and organic compound production, and in pollution monitoring. Few other organisms can boast such a successful record of usefulness to humanity! Four of the most important classes of life-saving pharmaceuticals known are produced by fungi: penicillin from Penicillium chrysogenum, cephalosporins from Acremonium chrysogenum, cyclosporin from Tolypocladium niveum, and lovastatin from Asperigillus terreus (Rossman 1997). Fungi are used in biotechnological processes or in the production of novel compounds, and they
Page 157
have a huge potential for use in the pharmaceutical and health-care industries (Fox 1993; Nisbet and Fox 1991; Rossman 1997; Wildman 1997).
Fungi can also cause huge losses of food in storage and substantial disease in crop plants in the field. Furthermore, because of their integral role in ecosystem processesfor example, in nutrient cycling, plant growth, as a food source, and in their sensitivity to air pollution and perturbationfungi (including lichen-forming species) are ideal organisms for measuring and monitoring biodiversity (Rossman 1994). Fungi have proved to be important, and it is up to mycologists to raise awareness of them among the wider public and politicians. Each mycologist has been challenged to devote a part of his or her working time to this task (Hawksworth 1995).
Numbers of Fungi
There are several estimates of the numbers of fungi (Cannon 1997a; Hawksworth 1991, 1993), and a working figure of 1–1.5 million species is now generally accepted (Hammond 1992; Heywood 1995; Rossman 1997). Several lines of evidence point to a similar figure, but it can be derived by comparing the number of fungi known in all habitats in a single geographical area (the British Isles) with the number of native and naturalized plant species in the same area (Hawksworth 1991). The resulting ratio of six fungi to each plant in an area, extrapolation to a conservative 270,000 global vascular plants, and the use of some allowances yielded a global total of about 1.5 million species of fungi. That figure contrasts markedly with the 72,000–100,000 species known (Hawksworth 1995; Hawksworth and Rossman 1997), and a few authors have argued that 1.5 million is too high (Aptroot 1997; May 1994); however, skepticism is based largely on a lack of familiarity with fungal distributions and host specificity and on the lack of detailed studies in the tropics. Recent studies in the tropics have found a magnitude of novelty that tends to support the figure of 1.5 million (Fröhlich and Hyde 1999; Hawksworth 1998; Hawksworth and Rossman 1997).
The extent to which new species are found varies among different systematic and ecological groups. For example, on the basis of the results of a monographic treatment of the saprobic ascomycete genus Didymosphaeria, Aptroot (1997) estimated that there were only 20,000–40,000 nonlichenized ascomycetes in the world. However, his estimate was based on the assumption that only seven of the 550 taxa classified in Didymosphaeria, actually belong to that genus (Aptroot 1995). He considered this a general trend in ascomycete systematics; although it might be for many long unrevised genera, a realistic figure has been used in totaling the world's described fungi (Hawksworth and others 1983, 1995). In other genera, the opposite trend is occurring. Oxydothis previously had 27 species names, but the number was increased to 42-after publication of the monograph of Hyde (1994), and a further 23 species have since been found on palms in Australia, Brunei, Ecuador, and Hong Kong (Fröhlich 1997). Aptroot's assumptions are also based on wide species concepts. How can we be sure that fungi with a wide host
Page 158
and biogeographical range and with varied structure are of the same species in the absence of inoculation experiments, incompatibility tests, and molecular data?
There are many potential pitfalls in endeavoring to extrapolate from limited datasets. For instance, a study of phyllachoraceous taxa in Australia and later extensive collecting across the continent increased the number of species known in the country only from 103 to 109 (Pearce and others 1997); extrapolation would provide lower estimates of fungi. But a study of palms in north Queensland identified 202 ascomycete taxa, of which eight genera and 95 species were new to science (Fröhlich and others 1997); extrapolation of these figures would provide much higher estimates of fungi. The original estimate of 1.5 million fungi (Hawksworth 1991) endeavored to account for numerous variables, and recent data from various sources (Cannon 1997; Fröhlich and others 1997a; Hawksworth 1993, 1998; Hyde 1995, 1996a) all point to the figure of 1.5 million as, if anything, conservative.
The number of vascular plants in the United Kingdom is about 2,089, and the number of fungi (including lichen-forming species) is estimated at 12,000 (Hawksworth 1991). Hong Kong, an island smaller than the Isle of Wight or Vancouver Island, has more than 1,700 vascular plants; if the ratio of six fungi to each plant species holds, there are more than 10,000 fungi in Hong Kong. We know of fewer than 500 species (or 5% of 10,000) of fungi in Hong Kong, but some plants have already been shown to support a large number of fungi, many of which are host-specific or family-specific (Fröhlich 1997; Fröhlich and Hyde 1999; Taylor 1997). Those findings indicate that a ratio of 1:6 for vascular plants to fungi might be low, at least in the tropics.
Measuring Fungal Biodiversity
Why should We Measure Fungal Diversity Rapidly?
The necessity for immediate assessments, new research, and rapid monitoring methods for measuring biodiversity is undisputed, and many of the general recommendations made also apply to fungi and other microorganisms (Burley and Gauld 1997). The ideal way to measure fungal diversity would be an all-taxa biodiversity inventory (Janzen and Hallwachs 1993) for fungian all-mycota biodiversity inventory (AMBI), as discussed further below. Because of the difficulty in detecting many fungi and because of their diverse nature, there is an urgent need for all fungi in one geographical region to be identified (Rossman 1994). This would provide basic data against which the results of other external and internal studies could be measured. However, because of the diverse ecologies, seasonality, sporadic findings, and so on, such a survey would take teams of specialists decades. There are at least 31 separate fungal niches in a tropical forest, almost all of which need different techniques and specialists to inventory (Hawksworth and others 1997).
Inasmuch as total inventories will always be impractical (except for a few sites), alternative methods for estimating fungal biodiversity and preparing environmental impact assessments are essential. If they can be developed, there will no longer
Page 159
be any reason for fungi not to be used widely in environmental monitoring, impact assessments, and ecological research. Indeed, because of their sheer diversity and niche specificity, fungi could prove to be especially valuable as indicators of different kinds of environmental changes and ecological processes.
An All-Mycota Biodiversity Inventory?
The need for an AMBI is undeniable. The scientific benefits would be immense with respect to providing a dataset against which to test hypotheses on species richness and host specificity. Permanent plots or otherwise circumscribed sites need to be established to initiate such an inventory. The most intensively inventoried sites for fungi in the world are two in the United Kingdom: Esher Common in Surrey and Slapton Key National Nature Reserve in Devon. Each has around 2,500 species recorded after several decades of work, but neither has been completely surveyedspecies are still being added, some niches have not been sampled, and the two sites have only about one-third of the recorded species in common despite many similarities in the plants of the two sites. The true number of fungi in the two UK sites, both of which have been intensively affected by human influences, could well be around 3,000. Whatever the total in those disturbed temperate sites, the richness in pristine tropical forests can be expected to be much greater because of the much larger numbers of potential host plants and insects. No sites in the tropics have yet been studied to a comparable depth; one was contemplated in Costa Rica but has since been abandoned, and some are now being planned in Brunei and Taiwan.
Biodiversity measurement in one plot or site is complicated by the need to sample large numbers of habitats and by the diversity of the fungi encountered. Most of the fungi collected can be predicted to be new to science, and their identification only to genus or family level might be possible (Hawksworth and others 1997; Hyde and Hawksworth 1997). In the case of the Guanacaste project, an estimated 50,000 fungi were probably present, of which around 35,000 could be expected to be new to science (Cannon 1996a). Once some site inventories are complete, protocols for accurate measurement of fungal diversity can be developed and tested in them. However, because of the problems mentioned above, it is unlikely that such results will be available within the next 20 years. In the interim, we must develop the best protocols we can on the basis of existing knowledge.
Alternative Approaches to Inventorying and Monitoring
What is our best way forward? The problems associated with selecting target genera or families or specific habitats as a measure of biodiversity have been discussed (Burley and Gauld 1996). The best approach is thought to be to integrate target groups and specific habitats. Carefully selected permanent plots (selected to incorporate a high degree of plant and habitat diversity) would be established under the auspices of local scientists. The plant species within a plot would be identified and labeled if possible. Mycological inventory could then be carried out over a period of years with input from appropriate specialists. The larger basidiomycetes (for example, polypores), ascomycetes (for example, Xylaria), and some biological groups of fungi (for example, entomophagous fungi, freshwater fungi,
Page 160
and lichen-forming fungi) could be collected and identified (spatially and temporally) over the whole plot, inasmuch as their numbers would be manageable. The microfungi could be investigated in smaller plots or individual host trees (Hyde and Hawksworth 1997).
Microhabitat Predictors
Because of the difficulties likely to be encountered if we choose to use predictor sets of fungi as a measure of species richness, another approach could be to select microhabitat predictors. Specific microhabitats in an area would be chosen, and a measure of the diversity of microfungi in those microhabitats would be made according to standard protocols. Random collections of leaf litter followed by isolation according to standard techniques might give us a good measure of overall diversity. If this were used with standardized isolations from random soil samples, estimation of fungal endophyte numbers in an endemic tree species, estimation of aerofungi and lichenized fungi on bark or leaves, and collections of Xylaria species, we might have a tangible, albeit qualitative, estimate of the actual diversity in a given region.
The microhabitat components chosen for such an approach to the estimation of biodiversity could vary from habitat to habitat, at this stage; we have no data on which microhabitats would be most representative of microfungal species richness. In the absence of an AMBI, a specific research effort on selected microhabitats is needed to assess which ones would yield the best indications of species richness, to prepare standard protocols for these, and to test the protocols for effectiveness, reproducibility, and ease of application. Five to eight years of coordinated effort across forest regions would be required to allow the identification of suitable predictor microhabitats and then to provide methods for rapid evaluations of fungal diversity in different regions.
Rapid Biodiversity Assessment
In rapid biodiversity assessment or RBA (Beattie and others 1993), numbers of fungi would be estimated without identification as to named species but by sorting them into recognizable, similar species units based on morphological similarities (Cannon 1997b; Hyde 1997; Hyde and Hawksworth 1997). Trained biodiversity technicians (“paratechnicians”) are required to sort specimens into recognizable taxonomic units (RTUs). It has been demonstrated that RTU estimates of spiders, ants, polychaetes, and mosses made by biodiversity technicians can be close enough to formal taxonomic estimates of species richness to be useful for RBA (Oliver and Beattie 1993), and we see no reason why this should not be tried with fungi.
The idea of applying RBA in mycology has been viewed optimistically by several authors (Cannon 1997b; Hyde 1997; Hyde and Hawksworth 1997), but it is not clear that workable reproducible protocols can be developed. High priority is now attached to the production of protocols, which can be tested by paratechnicians, revised, and widely promulgated, as addressed in more detail elsewhere (Will-Wolf and others 1999).
Page 161
Assessment with Molecular Techniques
The use of molecular techniques in estimating fungal biodiversity has been mentioned as a possibility (Cannon 1997b, 1999; Liew and others 1998) but not used. These techniques have been used to estimate bacterial species, including species that cannot be grown in culture (Tiedje and Zhou 1996), and theoretically they can be applied to fungi, although within-species genotype variation and the low proportion of fungi on which any sequence data are available, they pose particular problems in interpretation. Molecular techniques can be used to access litter and soil samples; although they are still tedious at the cloning stage, the rapid development of automated sequencing machines and computer generation of phylogenetic trees is making them increasingly feasible (Liew and others 1998).
Diversity Assessment with Image Analysis.
Computerized image analysis has been successfully developed to identify high-profile groups of fungi, such as airborne species, that might cause allergic responses in humans and trigger asthma (Benyon and others 1997). It could be feasible to develop identification by computerized image analysis for other groups of fungi, such as soil, litter, or mosaics of lichen-forming fungi on leaves or bark. However, computerized analysis is expensive to develop, and the method is unlikely to provide an alternative for wide-scale fungal assessments soon.
Selected Groups for Rapid Biodiversity Assessment
Recognizing the problems of inventorying all the fungi present in an area and the limitations of various other approaches as reviewed above, we discuss here some candidate groups for use in RBA.
Macromycetes
The large basidiomycetes are probably the easiest group of fungi to record in biodiversity surveys because they are conspicuous, easy to collect, and generally easily identified as to genus. Further separation at the species level can be carried out on site even if the fungi cannot be given an existing species name; they can be given numbers (for example, Coprinus sp.) (Hyde 1997). After a spell of rain, the fruiting bodies of macromycetes will flourish in most habitats, but it must be remembered that this is only a representative sample of those actually present; long-term studies over many years are needed to approach a full survey of larger fungi in a site, as demonstrated by studies in Malaysia and Puerto Rico in particular (Hawksworth 1993). Over a period of 7 weeks, sites in Tai Po Kau Nature Reserve and on the University of Hong Kong campus were visited during the wet season. Representative collections of all macromycetes were made on each visit and sorted into recognizable species units. The cumulative numbers of basidiomycetes at both sites indicated that the total numbers of recognizable species units had not been established. Numerous visits to each site are therefore required to obtain best estimates of species numbers.
Page 162
Xylariaceae
The Xylariaceae are a large family of ascomycetes, most of which are relatively conspicuous. They are particularly well represented in the tropics, although their identification at the species level might require good access to the literature, and some genera are rich in undescribed species in the tropics. They can easily be spotted in the field, occur on the forest floor, or sprout from dead stumps, logs, and branches; because of the robust nature of most species, they require minimal care in handling. A short visit to a site can generate large numbers of xylariaceous taxa. Identification to genus, species, or other recognizable units is relatively easy for paratechnicians. Because the fruiting bodies of these fungi are tough and persistent, they can provide a better comparative measure of fungal diversity than is the case with the more ephemeral, larger basidiomycetes.
Lichen-forming fungi
The lichen-forming fungi in tropical regions are confined mainly to the bark of trees and leaves. Their development depends heavily on light penetration, and in dense tropical forests most will be in the canopy layers. If they can be assessed, lichens can be especially attractive for RBA because of their perennial nature and variations in shape and color. In numerous cases, lichens have been surveyed by schoolchildren as a part of studies of air-pollution patterns, including one in Hong Kong (Thrower 1980). In tropical forests, some groups whose spores or other propagules are large or that for other reasons can be dispersed over only short distances (for example, Thelotremataceae) act as indicators of forests with long histories of ecological continuity; in Thailand, lichens on bark have been related to fire histories (Wolseley and others 1995).
The value of lichens living on leaves in the tropics as indicators of habitat disturbance has been demonstrated by a series of elegant studies in Costa Rica (Lücking 1997). The species forming mosaics on leaf surfaces lend themselves to being counted by eye and with a 10x hand lens by nonspecialists, so they can generate comparable data if similar trees and canopy-sampling strategies are used.
Lichens are now widely used in site assessments in temperate forests (Rose 1992) and merit parallel attention in the tropics. Lichens not only act as indicators of air pollutants and habitat disturbance themselves. Because a wide range of invertebrates feed on or are camouflaged to resemble lichens and provide hiding and breeding places for insects sought by insectivorous birds, sites with a high lichen diversity will also be rich in other dependent organism groups.
Endophytes
Endophytes are fungi or bacteria that for all or part of their life cycle live in tissues of living plants and cause unapparent and asymptomatic infections entirely in the plant tissues but cause no disease symptoms (Wilson 1995). There have been many papers on endophyte associations, mainly from temperate countries, but with some attention paid to tropical habitats (Dreyfuss and Petrini 1984; Fisher and others 1993; Rodrigues 1994; Rodrigues and Petrini 1997; Rodrigues and Samuels 1990). It is now believed that all plants have associated endophytes
Page 163
and that their foliage holds a reservoir of fungi, which can be easily recovered and isolated into culture. Isolation materials are widely available and inexpensive, and plant material is easily sampled and transported. Simple standardized protocols can be constructed to ensure comparability between samples, although allowances must be made for host specificity, and sampling is ideally restricted to particular kinds of trees.
Some endophytes have been shown to be organ- or tissue-specific, so sampling different parts of the plant (Fisher and Petrini 1988, 1990; Petrini and Fisher 1988) and varying the preparations and media used for their recovery yield different assemblages (Chapela and Boddy 1988; Fisher and others 1993; Petrini and others 1992; Pfenning 1997). With the exception of some fungi, such as xylariaceous anamorphs and some species of coprophilous fungi, endophytes are seldom recovered from soil or decaying vegetation (Bills and Polishook 1992).
J.E. Taylor has studied the endophytes and saprobes associated with the Chinese palm Trachycarpus fortunei, saprobes associated with Australian endemic Archontophoenix alexandrae, and the pantropical Cocos nuciferain and outside the natural biogeographic range of the former two species. Standardized sampling is needed at all the sites, and several sites at each location were investigated. Sampling was undertaken at the same time of the year (depending on seasonably and precipitation) to obtain comparable results. The results generated by both the endophytic and saprobic studies indicate a decrease in species numbers on palms when they are outside their natural range, unless they are in equivalent habitats within the case of palms, for instancea source of fungi from other palm hosts.
Selected Habitats for Rapid Biodiversity Assessment
As a complement to selecting particular groups of fungi to subject to RBA, we suggest that particular habitats be examined in addition to selected groups.
Palms
Palms are an integral part of most tropical forests and so are a valuable host group for comparisons of the fungi present. The fronds and stems of palms are robust, long-lived, and available for colonization by fungi over a relatively long period.
Investigations of palm pathogens, saprobes, and endophytes have revealed a high diversity of palm microfungi, mainly ascomycetes and related mitosporic fungi (Fröhlich 1992; Hyde 1992, 1993, 1994; Hyde and others 1997; Rodrigues 1994; Rodrigues and Petrini 1997). Fröhlich was intrigued by the seemingly limitless microfungal species that could be found on a single palm species in a given patch of forest and investigated the number of species that could be supported by a single host tree. An individual palm tree contains many distinct microhabitats: trunks, stems, roots, frond blades, petioles, inflorescences, fruits, seeds, and assorted appendages such as flagella and spines; these tissues vary in attractiveness to different fungi with age and health. To sample the mycota completely, it would be necessary to examine the following habitats separately:
Page 164
• all the living palm surfaces, especially the frond blade (phylloplane), for lichens;
• any diseased areas, such as leafspots or frond tips, for pathogens;
• the surfaces and interior of all the senescing and dead tissues for saprobes;
• living tissue collected in the field and incubated in the laboratory for latent pathogens and saprobes;
• the interior of all the healthy, fleshy organs, including the roots, for endophytes; and
• the root surfaces and interior for mycorrhiza.
Studies of the fungal saprobe numbers by Fröhlich and Hyde (1999) indicate that 172 species of saprobes occurred on three fan palms (Licuala sp.) in Brunei Darussalam (sampled three times over 1½ years), and 100 species of saprobes occurred on three fan palms (Licuala ramsayi) in Australia (sampled once). Palm saprobes could be a useful target group for biodiversity assessment; substantial data can be collected with minimal fieldwork.
Bamboo
Bamboo is also a good substrate for biodiversity assessment in the tropics because it is relatively common. In a preliminary study, a Bambusa sp. and Dendrocalamus sp. were collected in Tai Po Kau Country Park, Hong Kong, and on Mt. Makiling, Los Baños, in the Philippines. One decaying culm in each of three replicated clumps was cut down and chopped into pieces measuring about 25 × 3 cm. Twenty pieces were randomly selected and taken to the laboratory, where they were incubated and kept moist for 1–2 weeks. Each piece was microscopically examined, fungi were recorded, and the number of species on each host at each site was recorded.
The bamboo on Mt. Makiling was found to support 114 species, and that in Tai Po Kau Park 101 species. The hosts had different mycota, and the use of bamboo in RBA therefore seems likely to be effective.
Pandanus
Pandanus leaves are another good substrate for microfungi, particularly hyphomycetes. Collection is relatively simple and involves a pair of secateurs and thick protective gloves. Material can be collected dry or after rain. It should be returned to the laboratory and incubated for a few days. The hyphomycetes present on the material should sporulate quickly and can be identified to provide a measure of fungal diversity. Ascomycetes on Pandanus, bamboo, and palms can deteriorate after several days of incubation. If the material is allowed to air dry after a week of incubation, however, this arrests the deterioration of the ascomycetes and allows storage for long periods if necessary.
Freshwater Fungi.
Fungi flourish on submerged decaying plant material in freshwater; over 300 species of hyphomycetes (Goh and Hyde 1996) and about 300 species of ascomycetes (Shearer 1993) have been recorded. The number of new taxa is
Page 165
increasing (Goh and Hyde 1996). Thomas (1996) defines freshwater fungi as any species that rely on free freshwater for all or part of their life cycle. The richest fungal assemblages occur in average-size, more or less clean, well-aerated forest streams and rivulets with fairly turbulent water (Subramanian 1983).
The common sampling techniques involve collecting substratessuch as foam, water, and submerged plant debrisand examining them microscopically either directly or after incubation in moist or water aeration chambers. Plating is also common but is more labor-intensive and time-consuming and is not appropriate for rapid assessment. Foam filtration and water filtration are convenient and widely adopted. Foam and water samples usually contain conidia of numerous “Ingoldian fungi” with branched or coiled conidia that are separable microscopically without special training.
In contrast with the above methods, which yield mostly freshwater hyphomycetes, the incubation of wood samples from freshwater in moist chambers reveals diverse ascomycetes. Wong (1997) listed 363 species of freshwater ascomycetes, among which 303 were recorded on submerged wood, 18 on submerged bamboo, 40 on submerged leaves and two in foam samples. Hyde, Ho, Tsui, and Ranghoo, University of Hong Kong (pers. comm.), also noted that an extremely rich ascomycete biota occurred in tropical lakes and rivers. Examination of a good collection of wood samples by a mycologist takes about a month, in contrast with 3–5 days needed for a foam or water sample. However, it does reveal another important group of freshwater fungi, and it is therefore recommended for biodiversity assessment in freshwater habitats.
Pathogens
Plant pathogens might prove useful for estimating biodiversity. Collection will involve wandering around a site and collecting diseased leaves, which can be taken to the laboratory and examined. It is important that the collectors have a trained eye, but if this is the case it is possible to estimate diversity of plant pathogens in the field without laboratory examination. Many diseases are host-specific, and different fungal pathogens on a given plant usually differ in the symptoms that they cause.
It is rare to find leafspots on rain-forest plants, particularly palms and Pandanus species, and tar spots of phyllachoraceous taxa are also rare. However, large numbers of pathogens occur in gardens, nurseries, or monocultured crops.
Other Habitats for Rapid Biodiversity Assessment
The habitats suggested above are those we have worked on, and they have proved to be excellent sources of fungal diversity. Many others could perform equally as surrogates for biodiversity measurement of, for example, entomophagous fungi in the rain forests of Thailand (Hywel-Jones 1997) or leaf-litter fungi in Costa Rica (Bills and Polishook 1995). There are also ways of standardizing techniques for isolating soil fungi (Cannon 1996b). Disturbed forests harbor fewer rare species in the soil than undisturbed forest and so might be a good indicator of fungal diversity (Pfenning 1997). A concerted effort by mycologists is now needed to try to develop these target groups and microhabitat predictors. A given
Page 166
subset of target groups or microhabitat predictors is unlikely to work for all habitats, so a folio of basic methods must be selected to match local needs. However, until an all-mycota biodiversity inventory can be carried out, we would be wise to develop these methods to obtain estimates of fungal diversity.
Mycodiversity Technicians
It is unlikely that trained mycologists will always be available for or have the time to devote to measuring fungal diversity in a given habitat, but it can be possible to use mycodiversity technicians (Hyde 1997; Hyde and Hawksworth 1997). Mycodiversity technicians (a kind of “parataxonomist”) are not formally trained, but rather undergo minimal training to help in the task of biodiversity assessment. Their value can be exemplified by the use of students to measure endophyte diversity in one species of Pandanus and one of Livistona in Hong Kong and of summer students to measure larger basidiomycetes in two plots in Hong Kong.
In an experiment carried out with 54 students in Hong Kong, Livistona chinensis (a nonnative naturalized palm) and Pandanus furcatus were sampled for endophytes from the same piece of secondary woodland. The sampling and time-tabling were carried out as follows: Livistona chinensis (27 students divided into eight groups), eight plants sampled with 32 sampling units per individual (mature and immature leaves only); Pandanus furcatus (27 students divided into eight groups), eight plants sampled with 16 sampling units per individual (mature and immature leaves only). Far more fungi were recovered from P. furcatus in the pilot studies, so the number of sampling units had to be limited to a manageable amount. An alternative method would be to sample more individuals of a single host.
The general process was as follows:
|
Although some fungi would sporulate after 7 weeks, for the purposes of this practical class it was necessary to limit the number of weeks devoted to the study.
Several skilled demonstrators were necessary to assist with running the practical work, especially sorting the isolates into morphospecies and identifying them. In addition, the results were entered onto a database for the students, and the results were presented in a form suitable for statistical analysis. Assistance for 2 hours per week was also necessary for the intervening weeks when the students carried out subculturing. Recording of data was performed accurately, and there were few errors in the final dataset. The only technical problem was in numbering the individual isolates recovered by each group of students; this problem can
Page 167
be circumvented by allotting each group a series of numbers1–60, 61–120, and so onor different prefixes, such as A1–A60.
The study was labor-intensive and required considerable effort and input by the assisting demonstrators and technicians. However, the advantages were that this applied approach enabled students to investigate a fairly difficult concept and to carry out real scientific investigation on previously unstudied hosts. The students became proficient in sterile techniques, recording of results, and data analysis. The practical class can be carried out in later years on a variety of hosts, giving each group the chance to undertake a first study of endophytes from a specific host plant. Alternatively, technicians familiar with surface sterilization techniques and recording of data could undertake the labor-intensive parts of the work, leaving the identification to the trained mycologists.
In a separate experiment, we used four students over the summer break to compare basidiomycete diversity in a plot at Tai Po Kau Nature reserve with that in one on the Hong Kong University campus. The group first measured out a 1-ha plot and visited the site weekly for 7 weeks. During each visit, the students would walk through the plots, along paths parallel to one side of the plot; the paths were about 10 m apart. The mycodiversity technicians collected representatives of any macromycetes visible from the paths, placed them in suitable containers, and took them to the laboratory. The specimens were identified, isolated, photographed, and dried. Species that could not be identified were placed in recognizable taxonomic units and treated as above. Slide preparations were also made for future reference, and materials and slides are held in the herbarium (HKU) at the University of Hong Kong. Collections from later visits could be compared with photographs and slides from previous visits; in this way, it was possible to identify newly collected species.
Fifty-seven fungi were collected at Tai Po Kau Nature Reserve and 51 at the site at the University of Hong Kong. This indicates that the diversity of fungi was similar in the two sites. However, not all the macromycetes present would have been detected, so the cumulative number had not leveled off. It is interesting to note that more fungi were found at Tai Po Kau during the first visit (25 taxa) than at the other site (13 taxa). Although inconclusive, this pilot experiment indicates that
• mycodiversity technicians can be used in fungal diversity assessment;
• further studies are required to establish whether a single visit to a site to assess fungal diversity is representative; and
• further studies are required to establish how many visits to a site are required to collect an adequate representation of the macromycete species present.
Toward a Set of Protocols for the Rapid Assessment of Fungal Diversity
Here we propose protocols for several target groups and predictor habitats that should provide tangible estimates of fungal biodiversity in the tropics. We propose that at least six of these protocols be chosen to obtain a reasonable estimate
Page 168
of biodiversity and remove bias from any individual protocol. This set of procedures is proposed to provide a starting point for diversity assessment of fungi. The set can be tested, procedures can be added, and others can be removed, until we establish a robust mechanism for estimating fungal diversity across a range of global habitats. For the purposes of this exercise, we assume that appropriate human resources are not available and that specialized help will be provided by mycodiversity technicians. Most of the studies should be carried out during wet spells.
Macromycetes
The easily visible, larger fungi are an ideal target group on which to base biodiversity estimates, as long as they are integrated with other estimates to eliminate bias. We have found that a plot of 50–100 m2 can be thoroughly investigated in 2–3 hours, when all the larger fungi can be collected. This will require walking throughout the plot along lines 10 m apart, from where most fungi can be seen. Compartmentalized plastic fishing-tackle boxes or egg boxes are suitable and can be used to take the samples to the laboratory. It must be wet during the period under study, and at least 10 weekly visits should be made to each plot under investigation. It might be necessary to estimate the diversity of the longer-living polyspores on only one visit.
In the laboratory, untrained technicians can visually sort the material into morphospecies by using form and color. Slides and spore prints can be prepared, fresh specimens photographed, and single-spore isolations attempted. The specimens can then be freeze-dried or air-dried and placed in a reference collection for future study. Total diversity must exclude duplications of the same species collected on each visit. Over a 10-week period, it should be possible to obtain an indicative estimate of macromycete diversity by using mycodiversity technicians (mostly for unnamed specimens) or trained mycologists (for named specimens).
Lichen-Forming Fungi
Lichens are especially attractive for use in rapid biodiversity assessment because they are perennial and generally can be sorted into morphospecies first by eye and then with a hand lens. Although microscopic and chemical studies might be needed for critical determinations, they are not necessary when comparative assessments of species numbers are required. Experience with previously untrained students has shown that 1–2 days of training is sufficient to train a mycodiversity technician to survey these fungi. Lichens are long-lived, so only a single site visit is necessary, although ideally it should last for 2–4 days. The same sample plot as used for macromycetes could be surveyed, and it would be valuable to collect data on morphospecies distinguishable on tree bark and leaves separately. Where possible, canopy samples should be obtained from recently fallen or felled trees, although it is often the understory rather than the exposed crowns that are richest in leaf-inhabiting species. As many as possible should be examined because there can be variations due to light regimes and other microclimatic factors, which will affect the development of different species.
Page 169
The numbers of morphospecies can be compared directly by pooling or considering separately the datasets on bark and leaf-inhabiting species. The numbers of species to be found can be considerable. For example, studies of the crowns of 14 trees in a semideciduous tropical forest in Guyana yielded 100 lichen-forming fungi on leaves alone (Sipman 1997).
For those wishing to go further, the literature on the collection and identification of lichen-forming fungi is immense, but we recommend particularly a recent well-illustrated guide to New Zealand lichens (Malcolm and Galloway 1997). A detailed overview of lichen collection and identification is in press (Will-Wolf and others 1999).
Xylariaceae
The Xylariaceae constitute a tangible target group for biodiversity assessments because only a short period of training is needed to enable mycodiversity technicians to recognize them in the field. We have found that a plot of 50–100 m2 can be thoroughly investigated in 2–3 hours, when all the visible xylariaceous fungi can be collected. This requires walking throughout the plot along lines 10 m apart and closely examining potential substrates, especially logs on the ground. Most of these fungi are robust and require no special handling. It is often not possible to separate them in the field; therefore, all specimens will need to be returned to the laboratory for microscopic examination. It must be wet during the period under study, and we suggest that at least five visits, 2 weeks apart, be made to each plot under investigation.
In the laboratory, mycodiversity technicians can visually sort the fungi into groups (genera) according to form and to a lesser extent color. Further separations can be made with a hand lens or a dissecting microscope. Slide preparations of spores and asci are, however, essential for species-richness assessments, and the mycodiversity technicians will need to draw and measure these structures. Gross structure, asci, and spores could be photographed from fresh specimens, and isolations into culture from single ascospores can be attempted. The specimens can then be air-dried and placed in a reference collection for possible future study. Total diversity must exclude overlap of the same species collected on each visit. Over five visits, comparative estimates of the diversity of the Xylariaceae present can be obtained by mycodiversity technicians.
Logs and Branches
Numerous dead branches occur on the floor in most tropical habitats and can provide components for rapid fungal-diversity assessments. A short period of training provides mycodiversity technicians with the skill to collect logs and examine them for fungi in the laboratory. The plot of 50–100 m2 can be used, but in this case 20 logs can be randomly collected during a wet period and then incubated in moist chambers.
In the laboratory, mycodiversity technicians visually examine the logs with a dissecting microscope and make slides of fungi encountered. They can then sort the fungi into different groups on the basis of a minimum of taxonomic knowledge, that is, morphology, spore size, shape, septation, and color. In this way,
Page 170
fungi can be sorted into morphospecies. Photographs can be taken and specimens preserved or cultures attempted as for macromycetes. Total-diversity assessments must exclude overlap of the same species collected on each visit. We suggest that 20 samples is sufficient for mycodiversity technicians to provide a reasonably comparative estimate of the diversity of fungi on logs.
Endophytes
It is relatively easy for mycodiversity technicians to carry out standard procedures to recover endophytic fungi. The methods will depend on the host plant, and pilot studies need to be undertaken to develop optimal sampling and surface sterilization techniques. Surface sterilization techniques are outlined in the methodology of every paper dealing with the recovery of these fungi (Petrini 1986; Petrini and others 1992; Schulz and others 1994). The number of samples necessary to yield at least 80% of all the endophyte taxa at a single site has been estimated (Petrini and others 1992) at a maximum of 40 individuals per species and 30–40 sampling units per individual.
Although fairly labor-intensive, most of the techniques can be used by relatively unskilled technicians, or students, and results can be obtained in less than 3 months. The equipment necessary is inexpensive and widely available. Mycodiversity technicians will need to learn isolation techniques and spend some time at the microscope to separate fungi into “species units” or “morphospecies”. The same or allied hosts should be chosen to eliminate differences due to host diversity. Suggested species for which we already have results are palms (Fröhlich 1997; Taylor 1997), bamboo (Umali, unpublished), and mangroves (Rodrigues and Petrini 1997). Sporulation was promoted in many of these cultures with 43–52 species identified within 32 genera; however, different strains of the same species often exhibited different cultural characteristics.
Palm Fungi.
Palm rachids or petioles probably support the highest diversity of palm fungi, so we suggest these for biodiversity assessment. Numerous dead rachids can be found on the floor or attached to living palms in most tropical habitats and so are considered an ideal component of a suite of fungal-diversity assessment protocols. A short period of training will provide mycodiversity technicians with the skill to collect samples and examine them for fungi in the laboratory. The same plot of 50–100 m1 can be used, but in this case 20 rachid samples can be “selectively” randomly collected during a wet period or a dry period. They can be examined after air drying.
In the laboratory, mycodiversity technicians examine the samples with a dissecting microscope and make slides of fungi encountered. Fungi are sorted into different groups on the basis of minimal taxonomic knowledgespore type, spore size, shape, septation, and color. In this way fungi can be sorted into morphospecies. Photographs can be taken and specimens preserved or cultures attempted as for macromycetes. Total diversity must exclude overlap of the same species collected on each visit. We suggest that 20 samples are sufficient for microdiversity technicians to provide a comparative estimate of fungi on palms.
Page 171
Bamboo Fungi
Dead bamboo culms support a high diversity of fungi, and we suggest these for biodiversity assessment. Numerous dead culms can be found on the floor or standing, and two species of bamboo can be selected and sampled. Training, sampling, examination, and interpretation are similar to those for palms, and the same plot of 50–100 m2 can be used. Samples should first be examined for ascomycetes and basidiomycetes and then incubated for 14 days, after which they can be examined for other fungi.
Fungi on Pandanus
Dead Pandanus leaves support a high diversity of fungi, and we suggest these for biodiversity assessment. Numerous dead leaves can be found on the floor or attached to the plants, and these can be randomly collected. The type of training required and procedures to be followed are similar to those for palms and bamboos. The same plot of 50–100 m2 can be used, but in this case 20 leaves can be “selectively” randomly collected during a wet period or a dry period. Leaves should first be examined for ascomycetes and basidiomycetes and then incubated for 3 days, after which they can be examined for other fungi.
Freshwater Fungi
Comparative-biodiversity studies of fungi in freshwater habitats require the use of standard methods for foam and water examination. Examination of 10 foam collections and the filtrates from two membrane-filtered (pore size, 5–8 mm) 5-L water samples from three locations in the freshwater habitat should be carried out. Foam samples are collected in separated, clean, sterilized vials and preserved with the addition of formal-acet-alcohol or stored in an icebox. In the water-filtration method, the membrane filter is stained and fixed with lactic acid cotton blue or lactic acid fuchsin. This preserves and stains the spores and renders the membrane filter semitransparent. Samples should be examined until the number of new morphospecies recorded declines to a minimum; this can be assessed by plotting a cumulative graph of the number of new taxa recorded versus the number of slides examined in foam samples or the number of filter papers from the waterfiltration method examined. The direct examination of random wood samples for surface fungi could also provide a good estimate of fungal diversity.
Observation of conidia on semitransparent filter membranes might be difficult, especially with respect to the minute characters of fungal spores. The difficulty can be overcome with the use of a high-power dissecting microscope or a compound microscope with an upper light source. Conidia might also be covered by particles filtered in the filtration process; these can occlude important features.
Conclusions
Fungi are ideal organisms to work with in the field and in the laboratory. Collection requires a visit to the area under investigation and either collection of the visible fungi concerned or collection of small parts of the habitat under
Page 172
investigation. In the laboratory, the fungi are easy to handle and can be photographed and dried. Alternatively, isolations can be made with established techniques. Most fungi grow rapidly in culture and require no complicated procedures for their study.
We have endeavored to indicate the richness of tropical fungi and comparative approaches to the assessment of fungal diversity between different tropical sites. The several-protocol approach that we recommend is essential to capture some representation of the microfungi present. This is critical because the microfungi make up the highest proportion of fungi in any ecosystem. However, if time is short, we recognize that there could be advantages in paying particular attention to macromycetes and lichen-forming fungi in preliminary assessments.
In the exploration of fungal diversity, much attention has been focused on obtaining data on species richness in different systematic groups or in particular niches or substrates. Such studies have been important in vindicating hypotheses regarding the richness of the world's mycota, but we believe that it is now time to focus on securing comparative data on the richness of fungi in different sites. The approaches to rapid assessment of biodiversity in fungi described here are intended both to further discussion as to the best suite of protocols to recommend and more important to stimulate more work in tropical sites, even in the absence of experienced mycologists.
Acknowledgments
We are grateful to P.H. Raven for encouraging us to prepare this contribution for these proceedings. Helen Leung is thanked for technical assistance.
References
Aptroot A. 1995. A monograph of Didymosphaeria. Stud Mycol 37:1–175.
Aptroot A. 1997. Species diversity in tropical rainforest ascomycetes: lichenized versus non-lichenized; folicolous versus corticolous. Abstr Botan 21:37–44.
Beattie AJ, Majer JD, Oliver I. 1993. Rapid biodiversity assessment: a review. In: Beattie A (ed). Rapid biodiversity assessment, proceedings of the Biodiversity Assessment Workshop. Sydney Australia: Maquarie Univ. p 4–14.
Benyon FHL, Jones AS, Tovey ER. 1997. Image analysis differentiates spores of allergenic fungal genera and species. In: Barker RM, Barker WR (eds). Abstracts of the Joint National Conferences Adelaide '97. Adelaide Australia: Joint National Conferences Organising Committee, State Herbarium of South Australia. p 82.
Bills GF, Polishook JD. 1992. Recovery of endophytic fungi from Chamaecyparis thyroides. Sydowia 44:1–12.
Bills GF, Polishook JD. 1995. Abundance and diversity of microfungi in leaf litter of a lowland rainforest in Costa Rica. Mycologia 86:187–98.
Burley JM, Gauld I. 1996. Measuring and monitoring forest biodiversity: a commentary. In: Boyle TJB, Boontawee B (eds). Measuring and monitoring biodiversity of tropical and temperate forests. Bogor Australia: CIFOR. p 19–46.
Cannon PF. 1996a. An ATBIhow to find one and what to do with it. Inoculum 46(4):1–4.
Cannon PF. 1996b. Filamentous fungi. In: Hall GS (ed). Methods for the examination of organismal diversity in soils and sediments. Wallingford UK: CAB International. p 125–43.
Page 173
Cannon PF. 1997a. Diversity of the Phyllachoraceae with special reference to the tropics. In: Hyde KD (ed). Biodiversity of tropical fungi. Hong Kong: Hong Kong Univ Pr. p 255–278.
Cannon PF. 1997b. Options and strategies for rapid assessment of fungal diversity. Biod Cons 6:669–680.
Cannon PF. 1999. Options anc constraints in rapid biodiversity assessments in natural ecosystems. Fungal Div 2:1–15.
Chapela IH, Boddy L. 1988. Fungal colonisation of attached beech branches. II. Spatial and temporal organisation of communities arising from latent invaders in bark and functional sapwood under different moisture regimes. New Phytol 110:47–57.
Dreyfuss M, Petrini O. 1994. Further investigations on the occurrence and distribution of endophytic fungi in tropical plants. Bot Helv 94:33–40.
Fisher PJ, Petrini O. 1988. Tissue specificity by fungi endophytic in Ulex europaeus. Sydowia 40:46–50.
Fisher PJ, Petrini O. 1990. A comparative study of fungal endophytes in xylem and bark of Alnus species in England and Switzerland. Mycol Res 94:313–9.
Fisher PJ, Petrini O, Sutton BC. 1993. A comparative study of fungal endophytes in leaves, xylem, and bark of Eucalyptus in Australia and England. Sydowia 45:338–45.
Fox FM. 1993. Tropical fungi: their commercial potential. In: Isaac S, Frankland JC, Watling R, Whalley AJS (eds). Aspects of tropical mycology. Cambridge UK: Cambridge Univ Pr. p 253–63.
Fröhlich J. 1992. Fungi associated with foliar diseases of palms in North Queensland. Honours thesis, Univ of Melbourne.
Fröhlich J. 1997. Biodiversity of microfungi associated with tropical palms. PhD thesis, Univ of Hong Kong.
Fröhlich J, Hyde KD. 1999. Biodiversity of palm fungi in the tropics: are global fungal diversity estimates realistic? Biod Cons 8:977–1004.
Fröhlich J, Taylor JE, Hyde KD. 1997. Biodiversity of microfungi associated with palms in the tropics. In: Barker RM, Barker WR (eds). Abstracts of the Joint National Conferences Adelaide '97. Adelaide Australia: State Herbarium of South Australia. p 84.
Goh TK, Hyde KD. 1996. Biodiversity of freshwater fungi. J Indus Microbiol 17:328–45.
Hammond PM. 1992. Species inventory. In: Broombridge B (ed). Global biodiversity. London UK: Chapman & Hall. p 17–39.
Hawksworth DL. 1991. The fungal dimension of biodiversity: magnitude, significance, and conservation. Mycol Res 95:641–55.
Hawksworth DL. 1993. The tropical fungal biota: census, pertinence, prophylaxis, and prognosis. In: Isaac S, Frankland JC, Watling R, Whalley AJS (eds). Aspects of tropical mycology. Cambridge UK: Cambridge Univ Pr. p. 265–293.
Hawksworth DL. 1995. Challenges in mycology. Mycol Res 99:127–8.
Hawksworth DL. 1998. Getting to grips with fungal diversity in the tropics. In: Chou S (ed). Frontiers in biology: the challenges of biodiversity, biotechnology, and sustainable agriculture. Taipei Taiwan: Academia Sinica.
Hawksworth DL, Kirk PM, Sutton BC, Pegler DN. 1995. Ainsworth & Bisby's dictionary of the fungi, 8th edition. Wallingford UK: CAB International. p. 616.
Hawksworth DL, Minter DW, Kinsey GC, Cannon PF. 1997. Inventorying a tropical fungal biota: intensive and extensive approaches. In: Janardhanan KK, Rajendran C, Natarajan K, Hawksworth DL (eds). Tropical mycology. Enfield NH: Science Publ. p 29–50.
Hawksworth DL, Rossman AY. 1997. Where are all the undescribed fungi? Phytopathology 87:888–91.
Hawksworth DL, Sutton BC, Ainsworth GC. 1983. Ainsworth & Bisby's dictionary of the fungi, 7th edition. Kew UK: International Mycological Inst. p 445.
Heywood VH (ed.) 1995. Global biodiversity assessment. Cambridge UK: Cambridge Univ Pr. p 1140.
Hyde KD. 1992. Fungi from decaying intertidal fronds of Nypa fruticans, including three new genera and four new species. Botanical J Linnean Soc 110:95–110.
Hyde KD. 1993. Fungi from palms V. Phomatospora nypae sp. nov. and notes on marine fungi from Nypa fruticans in Malaysia. Sydowia 45:199–203.
Hyde KD. 1994. Fungi from palms. XII. Three new intertidal ascomycetes from submerged palm fronds. Sydowia 46:257–64.
Hyde KD. 1995. Fungi from palms. XIII. The genus Oxydothis, a revision. Sydowia 46:265–314.
Hyde KD. 1996a. Biodiversity of microfungi in North Queensland. Austral Syst Bot 9:261–71.
Page 174
Hyde KD. 1996b. Measuring biodiversity: diversity of fungi in the wet tropics of North Queensland. In: Boyle TJB, Boontawee B (eds). Measuring and monitoring biodiversity of tropical and temperate forests. Bogor Australia: CIFOR. p 271–286.
Hyde KD. 1997. Can we rapidly measure fungal diversity? Mycologist 11:176–8.
Hyde KD, Fröhlich J, Taylor JE. 1997. Diversity of ascomycetes on palms in the tropics. In: Hyde KD (ed). Biodiversity of tropical microfungi. Hong Kong: Hong Kong Univ Prp. p 141–56.
Hyde KD, Hawksworth DL. 1997. Measuring and monitoring the biodiversity of microfungi. In: Hyde KD (ed). Biodiversity of tropical microfungi. Hong Kong: Hong Kong Univ Pr. p 11–28.
Hywel-Jones NL. 1997. Biological diversity of invertebrate pathogenic fungi. In: Hyde KD (ed). Biodiversity of tropical microfungi. Hong Kong: Hong Kong Univ Pr. p 107–20.
Janzen DH, Hallwachs W. 1993. In: Highlights of the NSF-sponsored 'All taxa biodiversity workshop', Philadelphia PA. Bitnet, Biological Systematics Discussion list (taxacom@harvarda.bitnet) 16–18 April 93.
Liew EYC, Guo LD, Ranghoo VM, Goh TK, Hyde KD. 1998. Molecular approaches to assessing fungal diversity in the natural environment. Fung Div 1:1–20.
Lodge JD, Hawksworth DL, Ritchie BJ. 1996. Microbial diversity and tropical forest functioning. In: Orians G, Dirzo R, Cushman H (eds). Biodiversity and ecosytem processes in tropical forests. Berlin Germany: Springer. p 69–100.
Lücking R. 1997. The use of foliicolous lichens as bioindicators in the tropics, with special reference to the microclimate. Abstr Botan 21:99–116.
Malcom WM, Galloway DJ. 1997. New Zealand lichens: checklist, key, and glossary. Wellington New Zealand: Museum of New Zealand.
May M. 1994. Conceptual aspects of the quantification of the extent of biological diversity. Phil Trans R Soc London Biolog Sci 345:1–20.
Nisbet LJ, Fox FM. 1991. The importance of microbial diversity to biotechnology. In: Hawksworth DL (ed). The biodiversity of microorganisms and invertebrates: its role in sustainable agriculture. Wallingford UK: CAB International. p. 229–244.
Oliver I, Beattie AJ. 1993. A possible method for the rapid assessment of biodiversity. Cons Bio 7:562–8.
Pearce CA, Hyde KD, Taylor JE. 1997. Preliminary studies on the Australian phyllachoraceae. In: Barker RM, Barker WR (eds). Abstracts of the Joint National Conferences Adelaide '97. Adelaide Australia: Joint National Conferences Organising Committee, State Herbarium of South Australia. p. 87.
Petrini O. 1986. Taxonomy of endophytic fungi of aerial plant tissues. In: Fokkema NJ, Heuvel JVD (eds). Microbiology of the phyllosphere. Cambridge UK: Cambridge Univ Pr. p. 175–187.
Petrini O, Fisher PJ. 1988. A comparative study of fungal endophytes in xylem and whole stem of Pinus sylvestris and Fagus sylvatica. Trans Br Mycol Soc 91:233–8.
Petrini O, Sieber TN, Toti L, Vivet O. 1992. Ecology, metabolite production and substrate utilisation in endophytic fungi. Nat Tox 1:185–96.
Pfenning L. 1997. Soil and rhizosphere microfungi from Brazilian tropical forest ecosystems. In: Hyde KD (ed). Biodiversity of tropical microfungi. Hong Kong: Hong Kong Univ Pr. p 341–366.
Rodrigues KF. 1994. The foliar fungal endophytes of the Amazonian palm Euterpe oleracea. Mycologia 86:376–85.
Rodrigues KF, Petrini O. 1997. Biodiversity of endophytic fungi in tropical regions. In: Hyde KD (ed). Biodiversity of tropical microfungi. Hong Kong: Hong Kong Univ Pr. p 57–70.
Rodrigues KF, Samuels GJ. 1990. Preliminary study of endophytic fungi in a tropical palm. Mycol Res 94:827–30
Rose F. 1992. Temperate forest management: its effects on bryophyte and lichen floras and habitats. In: Bates JW, Farmer AM (eds). Bryophytes and lichens in a changing environment. Oxford UK: Clarendon Pr. p 211–233.
Rossman AY. 1994. A strategy for an all taxa inventory of fungal biodiversity. In: Peng CI, Chou CH (eds). Biodiversity and terrestrial ecosystems. Acad Sinica Mono Ser 14:169–94.
Rossman AY. 1997. Biodiversity of tropical microfungi: an overview. In: Hyde KD (ed). Biodiversity of tropical microfungi. Hong Kong: Hong Kong Univ Pr. p 1–10.
Schulz B, Wanke U, Draeger S, Aust HJ. 1994. Endophytes from herbaceous plants and shrubs: effectiveness of surface sterilization methods. Mycol Res 97:1447–50.
Shearer CA. 1993. The freshwater ascomycetes. Nova Hedwigia 56:1–33.
Page 175
Sipman HJM. 1997. Observations on the foliicolous lichen and bryophyte flora in the canopy of a semi-deciduous tropical forest. Abstr Botan 21:153–61.
Subramanian CV. 1983. Hyphomycetes: taxonomy and biology. London UK: Academic Pr. p 502.
Taylor JE. 1997. Biodiversity and distribution of microfungi on palms. PhD thesis, Univ Hong Kong.
Thomas K. 1996. Fresh water fungi. In: Orchard AE (ed). Fungi of Australia. Canberra Australia: Austral Biol Res Stud 1(B):1–37.
Thrower S. 1980. Air pollution and lichens in Hong Kong. Lichenologists 12:305–11.
Tiedje JM, Zhou J. 1996. Analysis of non-culturable bacteria. In: Hall GS (ed). Methods for the examination of organismal diversity in soils and sediments. Wallingford UK: CAB International. p 54–65.
Wildman HW. 1997. Potential of tropical microfungi within the pharmaceutical industry. In: Hyde KD (ed). Biodiversity of tropical microfungi. Hong Kong: Hong Kong Univ Pr. p 29–46.
Will-Wolf S, Hawksworth DL, McCune B, Sipman H. 1999. Assessing the biodiversity of lichenized fungi. In: Mueller GM, Bill G, Rossman A, Burdsall H (eds). Measuring and monitoring biodiversity: standard methods for fungi. Washington DC: Smithsonian Inst.
Wilson D. 1995. Endophytethe evolution of a term, and clarification of its use and definition. Oikos 73:274–6.
Wolseley PA, Moncreiff C, Aguirre-Hudson B. 1995. Lichens as indicators of environmental stability and change in the tropical forests of Thailand. Glob Ecol Biogeogr Let 4:116–23.
Wong SW. 1997. Ultrastructural and taxonomic studies of freshwater ascomyctes. PhD thesis. Univ of Hong Kong.
