Richard K. Benjamin, Meredith Blackwell, Ignacio H. Chapela, Richard A. Humber, Kevin G. Jones, Kier A. Klepzig, Robert W. Lichtwardt, David Malloch, Hiroaki Noda, Richard A. Roeper, Joseph W. Spatafora, Alexander Weir

Italics and other symbols are in the final version but not in this online copy!   Please send your comments by e-mail (btblac@unix1.sncc.lsu.edu) or if you have done lots of editing, by FAX (225-578-8459).

The numbers of some of these fungi are so large and their morphology so specialized only a few specialists study them. It is not surprising, therefore, that our knowledge of many species is so poor. This situation can be improved, however, only when they are better collected and better known, making their inclusion in biodiversity studies compelling.

Few protocols are available for quantitative sampling of insect fungi.  The problems involve the unpredictability of fruiting structures of some species, the microscopic nature of others and the need to culture them, and, most important, the patchy distribution of the associated arthropods.  In most cases sampling methods for the fungi must be targeted at the host insects. The protocols used will also depend on the condition (i.e., dead, dying, or alive) of the host and the stage (i.e., larval or adult) likely to support infection. For fungi found on living, adult hosts (e.g., Laboulbeniales and other taxa of unknown affinities) quantitatively gathered samples of potential hosts can be used to assess host utilization patterns and species richness of the fungi (Weir and Hammond 1997a, b). For these groups of fungi mass collections of hosts can be obtained by several trapping methods.

Flight interception traps
Large area flight interception traps about 10’ across consisting of black fine mesh netting suspended vertically above trays of preservative provide an effective means of obtaining quantitative samples rich in some groups of flying insects such as Coleoptera and Diptera, and these traps are not especially influenced by incidental variables.  Samples obtained in this way have been used to compare insect species richness in tropical and temperate forests (e.g., Hammond 1990) and to assess infection patterns of laboulbenialean fungi (Weir and Hammond 1997b).  Catches usually contain large numbers of predacious beetles, including Staphylinidae, and may also prove useful for bark and fungus-feeding species belonging to Scolytidae and Platypodidae.  In lowland tropical moist forest in Indonesia average weekly catches of beetles ranged from 100 to 250 species (Hammond 1990).

Light traps
Ultra-violet and mercury vapor lamps, usually used by lepidopterists, also attract a wide range of other insects including Coleoptera and Diptera.  Light traps frequently attract beetles associated with water such as Gyrinidae, Dytiscidae, and Hydrophilidae, and these can be a rich source of Laboulbeniales.   Between site and within site samples can be crudely quantified by trapping effort, and information on faunal (and fungal) differences with elevation can be assessed using transects.

Pitfall traps
Pitfall traps consist of plastic or metal containers sunk into the ground with the lip level with the soil surface. They provide a convenient means of trapping active ground-dwelling invertebrates such as carabid beetles.  The catch is determined by both population size and activity and is a measure of the "effective abundance" of the host (den Boer 1977).  Traps are usually placed in a grid system for predefined time periods, and the catches lend themselves well to mathematical manipulation (Luff et al. 1989).

Canopy fogging
The study of samples of insects obtained by insecticidal fogging has given an enormous insight into the structure and richness of the fauna of tropical and temperate forests. Individual trees are fogged from the ground using a synthetic pyrethroid, Reslin E. This is a nonresidual insecticide with high knock-down and low kill components. Arthropod samples are collected on both suspended trays (for statistical analysis) and on plastic sheets on the ground. Large numbers of beetles, especially Chrysomelidae and Corylophidae, hosts of species of Corylophomyces, Dimeromyces, Laboulbenia, and Rickia (Laboulbeniales) can be obtained in this way  (Weir and Hammond 1997b).

Litter samples
Known volumes of litter can be sampled for ground and litter-dwelling invertebrates. The litter must be gathered quickly to avoid losses and can be collected into large bags and processed through Berlese funnels on return to the laboratory. Litter samples can provide large numbers of beetles (principally Carabidae, Staphylindae, Pselaphidwaspae and Ptiliidae), ants, millipedes (Diplopoda) and mites (Acarina).

Other samples
Malaise traps provide another rich source of arthropod material, but comparison of catches can be difficult as trap composition is variable and dependent on the precise placement of the traps. For Trichomycetes associated with aquatic insects, it may prove possible to quantify catches using techniques already developed by river authorities for water quality assessment.
    Samples of dead and/or dying arthropods and their fungal associates are much more difficult to quantify, as they are usually collected by direct searching techniques. Nevertheless, sampling effort may be amenable to calibration and the results could be analyzed statistically.

Sites
The basic spatial unit of study for biodiversity investigations is the "site" and if sites are to be compared directly, it is important that a precisely equivalent sampling effort be undertaken at each site to eliminate or minimize bias. Ideally, surveys of insect fungi at a site should be carried out throughout the year, although little may be collected in temperate climates during the winter.

Collection
Once collected, fungi of interest or their insect hosts can be placed in a variety of containers in the field; each collector usually has her or his own preferences.  Substrate samples can be placed in paper bags, vials, or any other suitable containers.  Generally specimens should not be allowed to become too moist, and most collectors shun the use of plastic bags, although in some circumstances they may be preferred for the collection of aquatic samples.  If many small containers are to be used, it is convenient to have a single larger container to keep them together.  Baskets, fishing tackle boxes, and mesh bags such as those sold in diving shops are useful for this purpose.  Individual samples should be kept separate from others collected at the same time to avoid contamination.  Although often overlooked, it is important to use sterilized or at least new containers to prevent contamination from previous collections.  It is desirable to write and include field labels with the specimen at the time of collection.

Storage
Material collected from the field should be maintained in as unchanged a condition as possible until study, especially when it is to be cultured.  Although many fungi tolerate drying and can be cultivated from herbarium specimens many years old, others die within hours of drying.  When no information on drought tolerance is available and it is important to obtain a culture, material in fresh condition should be used. In some cases it may be necessary to attempt cultivation in the field.
    Refrigeration is a good way to keep some material in the laboratory for a few days; however, humidity in refrigerators often is very high and can lead to contamination by other fungi on the substrate. Walk-in refrigerators, in particular, can unintentionally become incubators containing a rich diversity of fungi.  Freezing of fresh specimens may be practical, even with living insects.  Care must be used, however, if the insects and fungi are to be kept alive, because many organisms, even those occurring in cold climates, seem to be relatively frost-intolerant.  Cooling or freezing may be a good way to slow down insects and mites for examination or to kill them for dissection.
    Some groups of insect-associated fungi never have been cultured.  Nevertheless, these fungi may well be of interest and certainly should be collected.  Most commonly fungi that are not to be cultivated are dried or preserved in various solutions before examination.  Large fleshy specimens are best dried on a dryer with a source of heat.  Smaller specimens can be dried in silica gel or simply air dried.  Oven drying is never recommended as specimens tend to cook.  Freeze drying is also possible, by lyophilization or even in a frost-free freezer.  When drying is undesirable, specimens can be preserved in solutions such as alcohol or formalin. Most solutions are not recommended, however, because nucleic acids or secondary products may be extracted over time.

Preparation of specimens for study
Specimen preparation is highly specialized for certain groups of organisms.  It usually requires mounting diagnostic parts of fungi from nature or cultures on microscope slides.  Part of the arthropod may be mounted as well.  Although certain stains usually are used for a mounting medium, we would almost always recommend that a water mount be made as well if living material is available.  In several cases when this was done, spores germinated in unexpected ways and provided more information on life histories of the fungi in question.  Also, all features of the mounted specimen are more natural and slight pigmentation can be seen easily.  Many mycologists insist that measurements be made in water mounts, but whatever the medium used for specimen measurements, this always should be stated.  Several general mounting media that may be used include lactophenol-cotton blue that has high affinity for fungal cells.  However, precautions should be taken when using phenol, because it is a carcinogen.  Glycerine jelly can be used as a mounting medium with water soluble stains, including cotton blue, and offers an alternative to lactophenol with the added value that it gels upon cooling.  Hardening mountants are available commercially, but the refractive properties may obscure some fungi.  Slides, even those prepared with aqueous mounting media, can be made to last over a hundred years if carefully prepared, and can serve as voucher specimens. Sandwiching a specimen between two glass cover slips is a particularly effective technique because it ensures that the sealant will fill the gap between the cover glasses and the environment.  These topics are discussed in more detail in sections to follow.
    Some of the fungi associated with arthropods are minute and have few characters.  Electron microscopic preparations may be necessary to observe certain structures, such as septal pores, to obtain phylogenetic information (Blackwell and Kimbrough 1976a,b).  Transmission electron microscopy has been of greatest value, but scanning electron microscopy can provide some information as well (Weir and Beakes 1996).
    More recently molecular techniques have been applied to arthropod-associated fungi to solve previously intractable taxonomic problems.  It is important to keep in mind that any collections made may be valuable for such studies.  For this reason it is best not to store specimens in liquid and not to dry them at too high a temperature for too long a time.  However, DNA can degrade when moist specimens lie about.

Cultivation and deposition of cultures
Many fungi are identified more readily from cultures.  Even when this is not so, it is advantageous to have cultures for physiological and genetic studies and for the production of secondary metabolites.  Dried cultures or those in a physiologically inactive state also may serve as type specimens for new taxa.  Media for specific kinds of fungi are given below.  A number of culture collections may accept cultures of arthropod-associated fungi.  These include the major general collections and more specialized collections such as the Collection of Entomopathogenic Fungi of the Agricultural Research Service (ARSEF) in Ithaca, New York (Humber 1992). If large numbers of cultures are to be deposited, however, it is best to inquire ahead about the willingness of the collection curator to accept them.  Unfortunately, not all groups of fungi associated with arthropods have been cultured, and sometimes it is necessary to maintain colonies of infected insects as a source of fungi that cannot be cultured.

Preparation and deposition of vouchers
Vouchers always should be prepared and maintained for all specimens of interest. Many of the fungi to be discussed are minute and are best preserved when mounted on permanent slides for deposition in a collection.  Dried cultures and usual herbarium specimens in packets or boxes may be prepared for other fungal groups.  Publications citing specimens that are not available to subsequent workers are of greatly diminished value compared to those with vouchered specimens.  Vouchers of insect hosts are equally important for obtaining identifications and document associations.  In both cases voucher specimens should exhibit the features that are used for identification.  Some care should be taken to deposit them in a collection that is likely to be well maintained for many years (Blackwell and Chapman 1993).  It is convenient for subsequent workers if specimens of a certain type are stored together in collections where there are similar specimens.  However, there are good arguments for not putting all specimens of rare organisms in a single collection, as many of those who have needed specimens from the Berlin Herbarium know.  Methods for the preparation of voucher specimens vary greatly from group to group. We discuss them in sections to follow.

Identification
Even when fairly up-to-date manuals are available, identification of fungi is difficult; however, the task is even more difficult when the literature is scattered as it so often is with this assemblage of organisms.  Identification by specialists is the most reliable way to name an organism, and nonspecialists must be prepared to pay for this essential service.  Identification services are offered by most major culture collections, as well as by many individual specialists.  If a collection service is to be used, it is advisable to select one with a specialist in the particular group of interest.  Your home institution mycologists should be able to help locate an appropriate specialist.  Those wishing to identify an unknown fungus on their own may find appropriate manuals for some groups of fungi cited in the references.  One reference has keys to families of the four fungal phyla and oomycetes (Hawksworth et al. 1995).  However, other groups mentioned here in passing, such as Myxomycetes and acrasid and dictyostelid slime molds, although studied by mycologists, are not included in the keys.  It also is important to identify the arthropod associate, and its identification at some taxonomic level at the very least will simplify identification of the fungus.  In fact it is much easier to have an arthropod specialist involved in a project from its inception.  Another way to discover diversity of some poorly-known fungal groups (such as Laboulbeniales) and avoid the problem of arthropod identification is to collect from specimens in well-curated collections where the identifications already have been made. 

Collection.  Ground-dwelling adult insects and larvae often are hosts of necrotrophic fungi, especially in years with high levels of precipitation at an appropriate season.  Many terrestrial necrotrophs of arthropods can be collected directly in the field by looking for dead or dying insects.  The insects may crawl up on living plants to which they become bound by hyphae of the parasite, or they may die hidden away in soil or under wood and stones (Keller and Zimmermann 1989).  In some cases healthy insects are associated with spores of necrotrophic fungi in their environment and become infected only under certain environmental conditions.  For this reason placing insects in a moist chamber may lead to infection in the laboratory.  This happens with subterranean termites, which often become infected with Conidiobolus coronatus in flooded moist chambers.
     Species of the chytrid genus Coelomomyces are notable as pathogens of mosquito or chironomid larvae; a second life cycle stage of each species parasitizes a copepod host.  These fungi can be found by collecting potential hosts.  For this purpose one may use a turkey baster as a syringe to draw water and larvae up for transfer into a plastic bag or other container.  The arthropod hosts should be held in the laboratory in shallow water in enameled pans or some other suitable container and examined over several weeks for signs of infection.  The fungal incidence may be low in nature, but generally is higher in laboratory situations.
     Entomopathogenic species of Cordyceps are not so frequently collected as other macrofungi (e.g., mushrooms), although, they are abundant in particular habitats.  Species of Cordyceps are found in the habitats of their hosts or in the specific habitats of the particular phases of the host life cycle that they parasitize.  The genus displays its highest species diversity in the tropics but occurs in other regions of high insect diversity as well.  The majority of species fruit during hot and humid seasons, and although phenology varies, there are exceptions to this generalization.  When searching for any fungus, including Cordyceps spp., an investigator must maintain a particular search image focus on appropriate and microhabitats.  For example, species of Cordyceps that parasitize subterranean larvae or pupae protrude from the soil.  Similarly, many species of Paecilomyces (an asexual state of Cordyceps) parasitize lepidopteran pupae that are found in the leaf litter or within decaying wood.  These fungi can best be located by focusing just above the leaf litter.  The stroma of the fungus will appear to be protruding from leaves on the forest floor, but actually it originates from a pupa on the underside of the leaf.  Other species that parasitize adults or nonsoil-dwelling phases of the host life cycle occur on leaves, stems, or other parts of living plants, which comprise different microhabitats to be searched.  Finally the plant community can be an important indicator of where to search, the hosts of Cordyceps are pollinators or pests of flowering plants.  An example of such a relationship are the more than 30 species of Cordyceps that seem to occur more frequently in rhododendron communities in the southern Appalachians of the United States, than in any other plant community.

Storage.  The most important step in preserving newly collected fungal pathogens of insects is to dry them quickly (by air-drying or using desiccants, very gentle heating, etc.) to suppress the saprobic bacteria and fungi or fungivorous invertebrates collected with specimens that all quickly overwhelm or destroy the desired pathogens. The spores of entomopathogens on properly dried specimens can remain viable for weeks or months, and can be cultured after returning to the laboratory. Entomophthoralean fungi are isolated most readily from very fresh specimens but, if dried before or during active production and discharge of spores, they may revive and continue sporulating when rehydrated later.  Because many insect necrotrophs can be cultured, we do not recommend preservation in alcohol except as a last resort.
     Fresh specimens of insect fungi must not be shipped in air- and water-tight containers unless the specimens are already quite dry. Whenever it is reasonable to do so, the specimens should be shipped in paper and cardboard containers that allow any remaining water vapor to escape; if a desiccant is included, it must be packed so that it will not damage a specimen with jostling during shipment.

Preparation of specimens for study. Specimens of most insect pathogens can be handled like other filamentous fungi and preparation of squash mounts usually suffices. However, some hard parts, such as sclerotia, may have to be sectioned with a freezing microtome or after routine fixation and embedding. Damage to the insect can be determined from sections, especially those that are plastic-embedded. The choice of mounting medium for slide making is rarely critical except to ensure consistent measurements. Some mycologists recommend making measurements from water-mounted material in which shrinkage does not occur.  Aceto-orcein is an outstanding routine choice for many insect pathogens because its high acidity serves to fix the fungus and wets the taxonomically critical structures of such common entomopathogens as Beauveria bassiana more easily than lactic acid-based mounting media.
     Aceto-orcein or other nuclear stains (e.g., Bismarck brown, methyl green, or aceto-carmine) may be required to identify fungi in Entomophthorales (Humber 1989). Family-specific differences in nuclear cytology are readily seen in unfixed fresh or preserved specimens.  Fungi in the Entomophthoraceae, many of which are necrotrophic parasites of insects, have large, readily stained nuclei with highly granular contents (Fig. 1). Fungi in the Ancylistaceae (Conidiobolus spp., few of which are entomogenous, have small nuclei that generally fail to stain in aceto-orcein (Fig. 2) because of the absence of condensed heterochromatin so prominent in nuclei of Entomophthoraceae.
 Slides, even those in which the fungus is in aqueous mounting media, can be made to last for many years by following one of the variant techniques of the double coverslip method (see "Preparation of specimens for study" below in the Laboulbeniales section for detailed protocol).

Cultivation and deposition of cultures.  The great majority of necrotrophic parasites can be cultured from conidium or ascospore inoculum on simple media.  Sabouraud dextrose agar + 1% yeast extract is very commonly used for these fungi.  The most fastidious fungi, especially some entomophthoralean species, may not grow in vitro from conidial inoculum; cultures of such fungi should be attempted using somatic inoculum obtained from surface-sterilized (and preferably living) infected hosts. In some entomophthoralean fungi, especially species of Entomophthoraceae, the somatic phase consists of protoplasts.  These fungi must be grown in more complex liquid culture media (e.g., Grace's insect tissue culture medium + 5-15 % fetal bovine serum), but usually will not sporulate in such a medium.  Humber (1994) discussed some of the problems of culturing and maintaining strictly obligate insect pathogenic fungi.
 Coelomomyces spp. present a bigger problem, but they can be reared with the mosquito and copepod hosts in the laboratory.  This process of course requires a constant source of uninfected laboratory-reared hosts.
 The United States Department of Agriculture (USDA) ARSEF collection, comprising about 5,000 isolates of more than 300 fungal taxa, is the world's largest and most comprehensive repository for cultures of insect fungi (Humber 1992).  Among the major general service culture collections, Centraalbureau voor Schimmelcultures (CBS, Netherlands) maintains many diverse insect fungi; the American Type Culture Collection (ATCC) also has some holdings of necrotrophic insect fungi. Both collections have online databases that can be searched by taxon or substrate (including host).

Preparation and deposition of vouchers.  Necrotrophic insect fungi are best deposited in herbaria as dried specimens preserved in packets or small boxes.  Storage containers prevent or minimize compression or mechanical damage of specimens. Specimens of infected hosts should be included with any stems, twigs, leaves, or other substrates to which they were attached when collected.  This is, obviously, not possible with specimens scraped off large, hard substrates (sound wood, stones, etc.) or dug up from soil, plant detritus, or rotting logs. Removing loose soil as well as mites or other mycophagous organisms from specimens before storage is essential.  We do not recommend that these fungi be preserved in alcohol unless no better (dried) alternative is possible.
 Major herbaria containing insect necrotrophic fungi include Kew and Commonwealth Agricultural Bureau International (CABI) (both with Tom Petch's extensive collections), Farlow Herbarium (with Thaxter's rich collections of Entomophthorales and Laboulbeniales), and the University of Michigan and University of Tennessee (with extensive collections of Cordyceps and related fungi by E. B. Mains and K. Kobayasi, respectively).

Identification. Identification aids for the majority of entomopathogenic fungi are spread through the literature and often are distressingly out of date, even for major taxa.  The most current and convenient general guide for insect fungi is by Samson et al. (1988).  Several extensive manuals for Entomophthorales (e.g., Keller 1987, 1991; Balazy 1993) and for Cordyceps (Kobayasi 1982) are available.  The indices in the ARSEF culture collection catalog (Humber 1992) offer the most comprehensive listings of fungal pathogens by their species, hosts, and collection localities.

Trichomycetes exist throughout the world, and in many regions they are common and abundant.  As with many other microscopic fungi, however, special techniques are needed to find, study, and culture them.  All species are associated with mandibulate arthropods that are detritivores, algivores, or ominivores, but apparently not with those that are predaceous, carnivorous, or that consume tissues of living vascular plants.  In almost all instances the fungi are hidden within the host's gut and not discernible until the animal is dissected in the laboratory and examined microscopically.  Currently, about 225 species and 55 genera of Trichomycetes are known (Table 3).  Six genera and more than 60 species have been discovered on various continents since Lichtwardt's 1986 monograph, attesting to the fact that the species richness of Trichomycetes is far greater than formerly realized. A supplement to the 1986 monograph covers the new genera and species (Misra and Lichtwardt  2000) .
 

Table 3.  Taxa, hosts, and habitats of Trichomycetes. Order Family      No. of  genera         No. of species               Hosts         Habitats Harpellales Harpellaceae a 33 >141b Dipterans larvae (mayflies, stoneflies) Aquatic (streams, ponds, pools) Asellariales Asellariaceae 3 11 Isopods, springtails Freshwater, terrestrial, or marine Eccrinales Eccrinaceae 14 52 Millipedes, crabs, anomurids, isopods, amphipods, insects Freshwater, terrestrial, or marine  Palavasciaceae 1 3 Isopods Marine  Parataeniellaceae 2 6 Isopods Terrestrial Amoebidialesc Amoebidiaceae 2 12 Larvae of insects or crustaceans Freshwater a Includes Legeriomycetaceae and Harpellaceae. b Includes new species not yet formally described. c Not phylogenetically related to Trichomycetes, but traditionally included in the class.

     Special adaptations that became established during trichomycete evolution have led to their success in living obligately within the gut of particular kinds of arthropods.  Although the fungi vary in the degree to which they infect populations of their hosts, in some cases infection of individuals may approach 100%.  This incidence is remarkable, considering that the fungi are shed at each molting event along with the linings of the gut to which they are attached, and that in most trichomycete species the number of reinfection propagules per thallus is significantly less than the number of spores produced by most other fungi, whether free-living or parasitic.
     Most species of Trichomycetes appear to be commensals and are innocuous, obtaining their nutrients from ingested substances passing through the gut.  There is some evidence that certain species of Smittium (Harpellales) may provide sterols and B-vitamins to mosquito larvae that are deprived of those essential nutrients. Determining such subtle nutritional relationships between fungus and arthropod is hampered, however, by our inability to culture most trichomycete species and the necessity that experimental hosts be free of other gut microorganisms.  At least one very widespread but apparently uncommon species, Smittium morbosum, kills mosquito larvae by inhibiting ecdysis.  It also has been demonstrated that some species of Harpellales in blackfly larvae occasionally grow from the gut into the developing ovaries, resulting in adult females that are sterile but can disseminate the fungus by flying to new sites and ovipositing the ovarian fungal cysts in place of eggs.  It is yet to be determined if this parasitic stage is a general means of dissemination in other Harpellales, all species of which grow and reproduce in nonflying larval forms.
     In Asellariales and Eccrinales, although immature host stages can be infected, full development of gut fungi occurs primarily in the sexually mature arthropod.  Because of several convergent similarities including host preferences, Amoebidiales have traditionally been studied by investigators of Trichomycetes, but evidence suggests that they are not closely related to the other three orders (Lichtwardt 1986).  None the less, Amoebidiales are included in this treatment because collectors of Trichomycetes often encounter them.

Collection and storage.  Trichomycetes live in a wide variety of hosts (Table 3).  It follows that the locations of gut fungi depend entirely on the habitats of their hosts.  In this section we provide a brief overview of techniques that can be used to obtain suitable arthropods; however, collectors should seek advice from appropriate specialists such as entomologists, benthologists, invertebrate zoologists, and marine biologists, especially those familiar with local faunas.  Lists of fungal species, their known hosts, and host habitats have been published by Lichtwardt (1986).
All Harpellales and Amoebidiales are aquatic, as are most Asellariales and some Eccrinales.  Harpellales and species of Paramoebidium (Amoebidiales) occur in the guts of insects that live mostly in lotic (flowing) waters.  The usual habitats are actively flowing streams, but can include edges of waterfalls and seeping cliffs.  Collecting in smaller streams is easier that collecting in large ones, and small streams contain often contain a greater diversity of larvae.  Harpellales are common in particular genera of mayfly and stonefly nymphs and in larvae of a number of lower dipteran families, such as nonbiting midges (Chironomidae), blackflies (Simuliidae), mosquitoes (Culicidae), and to a lesser extent in certain genera of craneflies (Tipulidae), biting midges (Ceratopogonidae), moth flies (Psychodidae), and solitary midges (Thaumaleidae).
    Lentic (still-water) insects in ponds, pools, lakes, and swamps, for example, mosquitoes, midges such as bloodworms, and a few other kinds of arthropods, may contain Harpellales.  A number of Asellariales and Eccrinales live in either lotic or lentic isopods and amphipods, as well as in some kinds of crayfish, freshwater crabs, hydrophilid beetles, and springtails (Collembola).  Species of Amoebidium (Amoebidiales) may be found on the exoskeleton of water fleas (Cladocera), mosquitoes, bloodworms, and crayfish.
    The most useful collecting instrument is an aquatic D-shaped net with a small mesh size.  Stream substrates consisting of rocks and gravel can be kicked with the feet while the net collects released insects that drift downstream.  Lifting larger rocks or scraping them with the hand often releases a variety of insects that can be caught in the net.  Many lotic insects prefer riffles and other agitated stretches of streams that are well aerated, although some seek zones where sediment collects.  Good sources of insects are vegetation, sticks, and small rocks.  These can be lifted from the water, and the attached insects harvested with blunt forceps. A woven-metal food strainer or sieve (12 cm diam) with a handle and with the support prongs bent backwards is useful in waters with abundant vegetation (borders of streams, marshes, etc.). Such strainers also can be used to sample muddy stream bottoms.
    Large, white plastic trays (about 40 x 30 cm), such as those used in photographic darkrooms, are useful repositories for the contents of nets and strainers or for materials plucked from the water.  The animals can be picked from trays with forceps or plastic droppers and placed into wide-mouth collecting jars with shallow layers of water.  Preferably all arthropods will be kept alive for dissection; consequently specimens should be kept cold in the field in an ice chest.  Some lotic insects may die soon after collection if they are not kept cold; others are hardier and will survive longer, provided the container with insects is kept in the shade.
    It almost always takes much longer to dissect, study, and process specimens than it does to collect them.  Depending on the species, aquatic insects can be kept alive for a day to several weeks after collection.  They should be refrigerated in shallow layers of water in an uncrowded condition in containers such as petri dishes or collecting jars.  Mosquito larvae, lentic bloodworms, and similar insects can be kept at room temperature.  It usually is best to separate the living specimens by type.  Predaceous insects that may have been collected should be discarded.
    Several genera of Eccrinales and one species of Asellariales (Asellaria ligiae) are marine (Hibbits 1978).  The fungi live in crustaceans that are mostly intertidal, or in the splash or high tide zone.  Hosts include isopods, amphipods, crabs, and anomurids such as hermit crabs and mud shrimps.  Some Eccrinales in galatheid anomurids are found below the low tide zone, even at abyssal depths (Arundinula abyssicola, around hydrothermal vents).  Crustaceans in deeper zones obviously require special equipment for collecting, but those that are intertidal or live along shorelines may be collected by hand or with nets at low tide.  Mud flats are home to several kinds of infected crabs (e.g., fiddler crabs) and anomurids that sometimes can be caught on the surface but may require digging with a trowel or shovel.
     Marine specimens can be placed in pails or other suitable containers, and must be kept from overheating but should not be refrigerated.  Shallow layers of seawater or damp seaweed in containers are satisfactory for transporting living specimens to the laboratory.  If the hosts are to be kept for a while before dissection and circulating seawater is not available in the laboratory, then small amounts of fresh water may be added to the seawater from time to time so that the seawater does not become too concentrated through evaporation.
    Terrestrial hosts of Asellariales and Eccrinales include millipedes, isopods, and a few kinds of beetles.  These can be collected by hand and should be placed in a container that is not tightly sealed, preferably with some of the host specimen's natural substrate.  Many terrestrial arthropods can be kept alive for long periods of time in a terrarium.  They need to be kept moist, but not too wet.

Preparation of specimens for study.  Dissection techniques are often a matter of individual preference.  In this section only basic methods for groups of arthropods will be presented.  Most dissections must be done under a dissecting microscope.  The most useful tools include two pairs of fine jeweler's forceps, a sharp single-edged razor blade, two very fine dissecting needles, and dissecting scissors, including a pair of fine iris scissors.
 Nymphs of mayflies and stoneflies can be grasped with jeweller's forceps at the posterior abdominal segment; when pulled, the hindgut is removed.  The epithelial layer can be stripped off with fine forceps in a drop of water, and the gut opened, if necessary, to reveal any Harpellales.  With dipteran larvae, such as midges and blackflies, either the hindgut and/or the midgut may contain Trichomycetes.  The posterior end and the head of such larvae can be cut off with a razor blade, and the gut removed.  Fungi in the midgut always are attached to the peritrophic membrane, a loose and transparent lining that is attached only at its anterior end.  The peritrophic membrane can be cleared of algae and debris by grasping one end and lifting it several times through the surface layer of the dissection water.  The hindgut epithelium should be removed to reveal Harpellales (or Paramoebidium) attached to the chitinous lining.
    The hindgut of amphipods also can be removed by pulling away the posterior segment of the body.  With isopods, the anal structures under the telson can be grasped with forceps and pulled.  Tearing apart in this fashion also can be used for removing the hindgut of some larger beetles and minute springtails (a special challenge!). Part of the exoskeleton of crustaceans such as crabs and anomurids usually must be cut away with scissors to reach the hindgut and stomach (foregut) where some eccrinids live.  The abdomen of crabs is folded under the animal; it can be pulled off and dissected to obtain the entire hindgut.  The simplest method for removing the hindgut of millipedes is to cut off the posterior end with a razor blade as well as about 1/4 of the anterior body.
    Host identification to the lowest taxonomic level possible is desirable, so voucher specimens should be preserved if the host is not already known.  Ethanol (70%) is generally a good preservative for most arthropods.  Undissected specimens are preferable for identification, but this is not always possible if few were collected. It is especially important to preserve the head of midge larvae, and the male genitalia may be necessary for identification of millipedes and some crustaceans.  Some aquatic dipteran larvae may be difficult, if not impossible, to identify to species.  For this reason, if pupae or adults are available in the field or emerge in the laboratory, they also should be made available to the specialist.
     Preserved arthropods can be used for microscopic examination, but this is not the method of choice in most instances, because dissection usually is more difficult, and artifacts almost always are introduced.  Recently dead specimens can be dissected, but trichomycete thalli tend to deteriorate very soon after the host dies.
     Water mounts on slides should be used for microscopic examination of most Harpellales, because if thalli are placed directly in fixative fine details such as trichospore appendages may be difficult, if not impossible, to discern properly.  In most cases hindguts can be torn open with fine needles or forceps and the thalli of harpellids spread out. The transparent peritrophic membranes need not be opened.  After identification, study, or photographing structures mounted in water, a partial drop of lactophenol with cotton blue can be placed on one edge of the coverslip and allowed to infiltrate as the water evaporates.  After sealing three sides of the coverslip with clear fingernail polish, the slide can be washed with water, dried, and then the fourth side sealed.
     Larger arthropods sometimes present a problem, because the chitinous lining with attached fungi often cannot be stripped off to reveal the thalli clearly under higher magnifications.  It may be possible to remove individual fungi in some cases or to pull off small pieces of the lining and mount them in water.  Eccrinales have unbranched, nonseptate thalli that are easily damaged.  The opened and washed gut of millipedes and some large crustaceans can be placed in dilute (10%) lactophenol for a few hours or overnight to loosen the chitinous lining.  Mounting larger pieces of the cuticle with many attached thalli often reveals several stages of development on one slide.

Cultivation and deposition of cultures.  Only some Harpellales (>190 isolates) and Amoebidium parasiticum (3+ isolates) currently exist in axenic culture.  The genera of Harpellales include Capniomyces, Furculomyces, Genistelloides, Simuliomyces, Smittium, and Trichozygospora.  Only Smittium, the largest genus, is represented by more than one species in culture; these currently consist of 13 named species plus many yet to be described.
     Some species of Harpellales culture rather easily; others require a good deal of persistence.  The perfered isolation medium is a dilute brain-heart infusion (1/10 BHIv):

Brain-heart infusion (Difco) 3.7 g
Thiamine HCl   200 Fg
Biotin    50 Fg
Glass-distilled water  1 liter
Agar    15 g

The two vitamins may not be essential, but there is some evidence that suggests thiamine is stimulatory to some species of Smittium.  This same medium can be used for storage of cultures in the refrigerator, and generally produces good sporulation.
    The following technique has proved successful in many isolations:  An insect larva is dissected and the hindgut removed.  It is not necessary to remove the outer layers of the gut, provided microscopic examination of a water mount indicates that a fungus is present.  The gut is washed at least twice in an antibiotic solution consisting of a stock solution of 40,000 units of penicillin G and 80,000 units of streptomycin sulfate per milliliter of distilled water.  This solution can be filter-sterilized directly into serum bottles and dispensed with a syringe.  Three to five drops of the concentrate is added to the wash water in 35 x 10 mm plastic petri dishes.  It is convenient to use a small loop (approximately 4 mm diam) to handle the specimen.
    After washing, the gut is transferred to a 60 x 15 mm petri dish with a thin layer of medium which has been overlayered with sterile, glass-distilled water, to which 3 to 5 drops of stock antibiotic solution have been added.  Most trichomycete species grow well at room temperature, but some, such as those from winter stoneflies (Capniidae), may have an optimum closer to 18° C. The culture must be monitored daily, using the low-power objective of a compound microscope.  If contamination appears, the specimen can be rewashed and replated.  Successful cultures usually will show growth in from 2 days to 2 weeks.  When growth is evident, the fungus can be transferred to medium without antibiotics, and later to a test tube slant containing a small amount of sterile distilled water (ca. 20 mm deep at the bottom of an upright tube) added after the agar gelled.  Until growth is well established, the tube should be rotated daily to allow some of the water to flow over the slant until eventually some of the fungus adheres to (but does not usually grow into) the agar.  In petri dishes, most harpellids produce many scattered colonies within the water overlayer, and colonies of some species may release trichospores in great abundance.
    Culturing Harpellales is different from culturing most other fungi: (1) water is necessary as an overlayer on agar medium; and (2) trichospores of most isolates do not extrude (germinate) in vitro, and consequently all transfers should be made by breaking up pieces of colonies with a loop.  Most harpellid isolates grow and sporulate well in shaken liquid culture, and the mycelium can be chopped with a sterile Waring blender, as with most other fungal cultures.
    Cultures of Amoebidium parasiticum can be started from pieces of the host to which thalli are externally attached.  Once growing, the small, unbranched thalli and sporangiospores are best transferred with a Pasteur pipette rather than with a loop.
     The preferred long-term storage method is in liquid nitrogen.  Trichomycetes do not withstand the lyophilization process.  Cultures can be maintained at refrigerator temperatures, and when thus stored should be transferred every 2 to 4 months, depending on the hardiness of the particular isolate.

Identification.  Morphological and other characters one needs to identify taxa of Trichomycetes can be obtained from descriptions, illustrations, and keys.  Knowing the type of host quickly narrows the possibilities of trichomycete identity, and in a few cases knowing the genus of the host provides identification of a known species.  Both sporulating and nonsporulating features are important in trichomycete identifications.  In all groups of Trichomycetes, as with most other fungi, it may be necessary to obtain measurements of reproductive structures and determine their shapes before identification is assured.  Excellent identification keys and biological information for all species of Trichomycetes are available (Lichtwardt 1986; Misra and Lichtwardt  2000).
    Taxa of Harpellales are identified primarily by thallus type (whether branched or not, amount and form of branching), basal (holdfast) structures, number of trichospore appendages, the presence or absence of a trichospore collar, and zygospore type. Thalli often are immature or devoid of zygospores; consequently preparations from several to many individuals may be necessary before trichospores or zygospores are found.  Appendages appear in trichospores only after their release from generative cells.  If maturing trichospores on sporulating branchlets are present, keeping the slide with the water-mounted specimen in a moist chamber for several hours to overnight may result in the release of some trichospores.  Occasionally, zygospores may mature under the same conditions.
     In Asellariales, the holdfast structure is especially important in the identification of species.  The holdfast of Eccrinales is a useful character, but more emphasis is placed on the shape and size of the thallus and the various types of sporangiospores produced.  Many genera and species of these orders are easily identified.  Others require many preparations before identification is assured.

Laboulbeniales
Laboulbeniales is a distinctive group of obligately biotrophic parasitic ascomycetes that lack mycelium.  They live on a diverse group of arthropods.  Most species grow on true insects (Hexapoda) and are known on the following orders: Cursoria (Blattaria), Coleoptera, Dermaptera, Diptera, Heteroptera, Hymenoptera, Isoptera, Mallophaga, Orthoptera, and Thysanoptera.  Relatively few (54) species infest mites (Class Arachnoidea; Order Acarina) and millipedes (Class Diplopoda, Order Juliformia).  None has completed its life cycle, i.e., produced ascospores, in axenic culture.  The discussion in this section pertains to species classified in Ceratomycetaceae, Herpomycetaceae, Euceratomycetaceae, and Laboulbeniaceae. Currently, Tavares' (1985) treatise, Laboulbeniales (Fungi, Ascomycetes), is the most comprehensive source of general information available on families and genera, development, morphology, sexuality, and distribution of Laboulbeniales.  This work is a necessity for anyone interested in the systematics of these fungi.  Equally important is the classic, beautifully illustrated, monograph of Thaxter, Contribution Towards a Monograph of the Laboulbeniaceae (Thaxter 1896, 1908, 1924, 1926, 1931).  A supplement to Thaxter's work (Benjamin 1971) also is an essential aid to the study of the group.  Regional studies that offer much useful information on ecology, general biology, collection, and preparation of specimens, as well as taxonomy, include those of Huldén (1983), Majewski (1994), and SantaMaría (1989).
     Because the thalli of the group are so different from those of other fungi, it is important to consider their morphology.  All Laboulbeniales are relatively small, ranging in length from about 50 mm to 1mm.  The fungal thallus develops directly from a germinating ascospore, which may undergo a precise sequence of cellular divisions, at least during the early stages of growth.  The main part of the body of the young thallus is termed the receptacle and consists of few to many cells often arranged in a particular order.
     The receptacle is attached to the host by its modified basal cell, or foot, from which a simple or sometimes branched haustorium develops.  Haustoria usually penetrate the host no farther than the living cells of the epidermis; in some species, however, haustoria are less localized and may penetrate and even ramify some distance into the body cavity (Thaxter 1896, 1908, 1924, 1926, 1931; Tavares 1985).  Laboulbeniales appear not to be pathogenic, and evidence suggests that they cause little, if any, damage to their hosts.
     The cells of the receptacle may bear simple or branched, often several-celled, appendages. The appendages may be sterile or fertile.  The latter produce minute, uninucleate, nonmotile cells, the spermatia, which are assumed to have a sexual function.  In the Ceratomycetaceae, tiny branchlike spermatia appear to develop directly from the cells of an appendage.  In the Herpomycetaceae, Euceratomycetaceae, and Laboulbeniaceae, spermatia develop inside distinctive structures termed antheridia.  Antheridia may be intercalary cells of an appendage; simple, free phialides that discharge spermatia directly to the outside; or more or less complex assemblages of closely associated fertile cells that discharge spermatia, into a common chamber from which they escape to the outside via a single opening. As the thallus develops, it gives rise to one or more perithecia. Each immature perithecium gives rise to a female receptive structure, the trichogyne. Spermatia appear to be transferred passively from antheridium to trichogyne. Actual fusion of sexual nuclei has never been observed, but it is presumed to occur. In any event, a centrum (the perithecial contents), develops within the maturing perithecium and forms one or more ascogenous cells that produce a succession of asci containing usually four ascospores (Tavares 1985).  Ascospores of all known Laboulbeniales are more or less acicular and two-celled.
     Of the 137 genera of Laboulbeniales currently recognized, 120 appear to be monoecious, the other 17 are exclusively dioecious or include dioecious as well as monoecious species.  Two species of Triceromyces, which have both monoecious and dioecious morphs, represent the only known examples of apparent tridioecism in the fungi (Benjamin 1986).  Dioecism apparently has arisen several times in the order.  In some genera, for example, Dimeromyces, Dimorphomyces, Trenomyces, and Laboulbenia, which has only a few dioecious species out of many hundreds, males and females are morphologically similar except for the production of sexual organs.  In other genera, for example, Amorphomyces, Dioicomyces, Aporomyces, Corylophomyces, andRhizopodomyces, the male may be reduced to a single series of two or more cells bearing a terminal antheridium.  In Aporomyces and Dioicomyces the ascospores giving rise to the two sexes may be strongly dimorphic, those of the male often being greatly reduced in size compared to those of the female (Benjamin 1989).

Collection.  Field collection of Laboulbeniales depends on collection of the hosts.  Few of the thalli can be seen well in the field, and the success of a field trip truly can be judged only after microscopic examination of the insects. The Laboulbeniales parasitize many groups of true insects. Beetles (Coleoptera) and flies (Diptera) are hosts of a number of cosmopolitan species and are relatively easy to collect. Staphylinidae harbor species of many genera (e.g., Corethromyces, Monoicomyces, Rhachomyces, Teratomyces); carabid beetles are hosts of many species of Laboulbenia; and flies, to over 100 species of Stigmatomyces.
     Likely hosts for Laboulbeniales live in a wide variety of habitats: water, soil, decomposing plant and animal remains of all kinds, flowers, stems, and foliage of living plants, as well as on the bodies of living animals such as bats and birds.  Collecting insects from many such habitats may call for specialized techniques, which can be found in the entomological literature pertaining to given groups.
     In tropical or subtropical regions where insects may be active throughout the year, collecting Laboulbeniales may be profitable at any time, being influenced primarily by whether the season is wet or dry.  In northern or southern climes where winters may be severe and insects more or less inactive until the return of clement weather. In such areas collecting should be best in spring or early fall when the degree of infection of host populations may peak.  Often, only a few individuals in a given population of an insect may be infected with fungi.  Thus, the goal always should be to make mass collections of a variety of hosts to assure success in obtaining a variety of parasites.
     Some equipment that is not generally carried about by mycologists, can make collecting Laboulbeniales much easier and more profitable. Such items include heavy gloves; forceps with both coarse and fine points for handling living or dead insects; a knife, trowel, or other strong tool for stripping bark, breaking open rotten logs and stumps, and digging in soil or detritus of all kinds; assorted plastic vials having screw caps with tight seals; a small funnel; a hand lens; a plastic bottle of preservative (70% alcohol); pencils; paper for labels, and a notebook.  A pump-spray can of insect repellent (30% Deet or more) not only makes collecting in the haunts of insects more comfortable, but also is especially useful for killing flies or other insects collected in a net.  Specialized equipment needed for certain insects includes a deep, flat-bottomed insect net constructed of light-weight nylon cloth for capturing flies and other insects on the wing, one or two small nets for capturing aquatic insects (those sold by dealers in tropical fish are excellent), and an aspirator for capturing terrestrial insects.  An aspirator is easily made as follows from a glass or, preferably, plastic bottle (a 50 cc centrifuge tube is excellent, Walter Rossi, pers. comm.) having a screw-cap lid:  (a) Two holes having diameters of ca. 7 mm and 10 mm, are drilled through the lid a few mm apart.  (b) Ends of flexible vinyl tubing of appropriate diameter and ca. 30 cm long are inserted in the holes for a distance of ca. 1 to 1.5 cm.  (c) A bit of fine-mesh cloth affixed to the short end of the small-diameter tube prevents insects or detritus from moving upward when the collector sucks on the long end while holding the long end of the large-diameter tube close to but not quite touching a desired insect.  The insect will be drawn into the bottle along with the inrushing air.
     A sifter to aid in separating insects from ground litter, flood debris, and detritus of all kinds can be made of a square or rectangular piece of hardware cloth having 8, 10, or 12 meshes to the inch.  Material to be checked for insects is placed on the sifter and shaken over oilcloth or a plastic pan with more or less vertical sides.  Insects falling through the screen are captured using the aspirator.
     A Berlese funnel set up in the laboratory is ideal for recovering large numbers of arthropods from materials returned from the field.  This consists of a large funnel, ca. 10 to 12 in. in diameter at the top with a circle of hardware cloth similar to that used for the sifter secured ca. one-third of the way down the funnel.  The funnel is suspended by a ring stand or other support with the small end inserted into a container of preservative such as 70% ethyl alcohol (added after the funnel has been charged with material to be examined) for catching insects and other arthropods falling from debris placed on the screen and covered with a cloth or plastic sheet held in place by a strong rubber band.  Heat, not so intense that it kills the insects, can be applied above the retaining cloth (an electric light with suitable reflector).  As the debris in the funnel dries out top to bottom, insects migrate downward and fall into the preservative from which they can be recovered.
     Beetles (e.g., Hydrophilidae, Dytiscidae, Haliplidae, and Gyrinidae) living in water are best captured using small fish nets or tea strainers.  Others (e.g., Carabidae and Staphylinidae) that live under stones, logs, wood fragments, and piles of flood debris are most easily collected using a sifter and aspirator.  Many insects live on or in mud or sand; these materials as well as other debris can be immersed in near-shore water; insects floating to the surface are netted and picked out with tweezers.  Members of several families of Diptera that are hosts of Laboulbeniales, especially Ephydridae and Sphaeroceridae, frequent mud flats at the margins of streams, lakes, and ponds; these are best captured with an aspirator or by sweeping with the net.  A very brief application of insect repellent to the net will stun or kill the flies (and other insects), which can be removed with forceps and transferred to alcohol. Rich communities of infected beetles can be found under algal drifts on the coast or under general debris around the margins of reservoirs. In those environments beetles are more or less confined to linear habitat strips, and levels of infestation by Laboulbeniales and other fungi can be high (approaching 100%) (P. M. Hammond and A. Weir, unpubl. data).
     Leaf mold on the forest floor, accumulations of rotting vegetation, such as grass piles and garden debris, and litter in hollow trees and stumps are inhabited by many kinds of insects and other arthropods.  Insects living in such materials (e.g., Anthicidae, Carabidae, Staphylinidae) are best collected by sifting in the field, or by processing samples in a Berlese funnel in the laboratory. Other forest insects can be caught in flight interception, light, or pitfall traps, or, for the more specialized collector, by canopy fogging. Fruiting bodies of larger fungi also support a wide range of invertebrates that can be collected using an aspirator or tweezers.
     Material obtained by sweeping of emergent vegetation with an insect net or beating and shaking branches and flowers can be emptied into a large pan, a cloth spread on the ground, or a beating sheet (a square of cloth stretched taut between the ends of a pair of crossed sticks or thin boards).  The dislodged insects can be caught with an aspirator and transferred to alcohol.
     Insects collected in the field are best preserved in 70% ethyl acohol for transport to the laboratory.  Transfers from aspirator bottles to vials should be made frequently to reduce the chance of mechanical damage to fungi and hosts by debris that inevitably is drawn into the aspirator along with the insects.  Single insects can be transferred to vials by means of forceps.  Those captured in quantity with the aspirator can be transferred to a vial of alcohol with the aid of a small funnel.

Preparation of specimens for study.  Removal of the thalli of laboulbenialean fungi to a slide mount is a challenge eased somewhat by the proper equipment and supplies.  Tools and other equipment include:  (1) High-quality, stainless steel, watchmaker's tweezers, which can be kept sharp pointed using a fine-grained grinding stone.  These are essential for sorting hosts and manipulating insects when preparing slide mounts.  (2) Several sizes of flat-bottomed porcelain imbedding dishes or other suitable containers for sorting and examining insects in alcohol.  (3) Ordinary depression microscope slides for temporary storage in glycerol of parasites removed from a host as well as large-capacity Maximov depression slides having a concavity ca. 35 mm in diameter and 5 mm deep in which to manipulate insects in glycerol when removing parasites.  (4) Stainless steel Minuten insect pins (available from entomology supply companies) mounted in the end of a match stick or held in a pin vice for detaching parasites from insects.  The pins can be kept sharply pointed using the grinding stone.  (5) Microscope slides (1 X 3 inches) and both 22 and 18mm cover glasses (#0 or #1), either square or round.
     Mounting media used include:  (1) Aqueous glycerol (glycerol, 100 ml; distilled water, 5 ml; chloral hydrate, 5 g), with or without a trace of a dye such as cotton blue or acid fuchsin, is the preferred medium for mounting Laboulbeniales for general morphological studies.  Its refractive properties are such that the relationships of cells comprising the thallus are observed easily.  (2) Amann's solution (phenol crystals, 20 g; lactic acid, 16.5 ml; glycerol, 32 ml; distilled water, 20 ml) can be used with or without a trace of acid fuchsin or cotton blue, and the solution can be substituted for aqueous glycerol.  (3) Hoyer's medium (gum arabic, 30 g; chloral hydrate, 200 g; glycerol, 16 ml; distilled water, 50 ml: the gum arabic and chloral hydrate, in turn, are dissolved in the water without heat, added to the glycerol and mixed thoroughly).  Use sparingly only as a medium in which to position fungi on slides or cover glasses, before adding mountant.
     As already stressed, insects collected in the field or by means of a Berlese funnel in the laboratory should be stored in 70% ethyl alcohol.  Gross collections usually will be contaminated with dirt or other debris. The insects should be segregated from such material with the aid of a dissecting microscope and fine tweezers and transferred to fresh alcohol before being examined for fungi. All parts of the body of each insect is examined carefully under medium or high magnification.  In some cases it is helpful to have the insects in a glass container with light directed from below to show the thalli in profile.  Parasites are sometimes relatively numerous, more or less generally distributed on the surface of the host, and easily found. Others may be limited in number and/or distribution and more difficult to detect.  The color of some may contrast sharply with that of the host whereas others may be nearly concolorous.  Some may be more or less erect, others closely appressed to the host integument.  Insects often are covered with setae that tend to obscure or be confused with the parasites.  Most Laboubeniales have a blackened foot that may contrast in color with that of the rest of the ascoma or insect and can reveal the presence of a parasite on the host's integument.  Parasitized insects should be stored in alcohol along with a label giving complete collection data.
     Each parasitized host is examined carefully under a dissecting microscope so that the exact locations of parasites on the host's body, either scattered over the surface or concentrated in clumps of few to many individuals can be recorded. Some species of Laboulbeniales can grow anywhere on the host body, whereas others are highly specific in this regard.  Thus, groups of closely associated thalli could represent several distinct species. Specimens from different groups should be mounted on separate slides, preferably several slides for each group. Each group is sampled separately. As an additional precaution against mixing species, the insect and the micropin should be rinsed with alcohol after preparing slides from one group before proceeding to the next group.  Some species of Laboulbeniales are dioecious, and the males may be extremely small, often little larger than an ascospore. Their detection will require special effort.
     Because someone may wish to study development of a given species, immature as well as mature specimens should always be mounted if possible.  Usually, development involves several kinds of transitory stages many of which may have great value in assessing relationships among species and genera. Examples, include antheridia, which often do not persist in mature ascomata, and trichogynes, which typically degenerate soon after perithecium development begins.
     The best procedure for preparing slide mounts is a slight modification of the double cover glass method adopted some years ago by Benjamin (1986: 247-248; see also Chupp 1940 and Kohlmeyer and Kohlmeyer 1972): (1) An infected insect is placed on the stage of a dissecting microscope immersed in plain glycerol solution in a large concavity slide. The viscosity of the solution facilitates manipulating the insect and the fungi to be removed from it.  (2) While the insect is grasped with fine forceps held in the nondexterous hand, the parasites are carefully detached with a Minuten needle. Great care must be used to avoid damaging the insects and fungi.  (3) The fungi are immediately mounted or stored temporarily in a small drop of glycerol in a depression slide.  (4) A minute amount of Hoyer's medium is dropped on the center of a 22-mm cover glass on a microscope slide (squares cut from gummed labels affixed to the slide right and left of the cover glass help maintain its position).  (5) The depression slide carrying the fungus and the slide carrying the cover glass are placed side by side on the stage of a dissecting microscope and, with the nondextrous hand, are moved backward and forward as needed while the specimens are transferred with the dexterous hand.  (6) The fungal specimens are positioned with the receptacle directed upwards so that the thallus will appear erect when viewed with the compound microscope.  (7) A few fibers of lens tissue placed adjacent to the fungal specimens will help to prevent their being flattened or distorted when the second cover glass is added.  (8) A small drop of mountant is centered on an 18-mm cover glass and carefully inverted over the specimens on the 22-mm cover glass. Mountant is best dispensed using a squeeze-type plastic dropper bottle having a tiny orifice. When placed on the cover glass, the drop should spread to a diameter of no more than ca. 3 to 4 mm. Only enough mountant to fill the space between the two cover glasses is needed. (9) Another small drop of clear glycerine is placed in the center of the 18-mm cover glass.  (10) Then, without reversing, the two-cover glass combination is inverted onto on a clean microscope slide in a position sufficiently offset to allow a suitable label to be attached to the slide. The glycerol should reach the edge of the small cover glass, which is now in direct contact with the microscope slide, and will help prevent inwelling of the sealing compound.  (11) The large cover glass, now uppermost, is sealed to the microscope slide by spreading (on opposite edges of one side of the cover glass) an excess of a medium such as Canada balsam or Fisher Permount7, which harden in time. The sealant will dry more quickly if the slides are placed on a slide warmer set at approximately 40*C for several days.  The heat also hastens diffusion of the Hoyer's medium surrounding the fungi into the glycerol and the plumping of the specimens.  During the first day or two, while the sealant hardens around the edges of the large cover glass, it is advisable to place a small weight on the cover glass. Small metal nuts weighing ca. 0.5 g each are available at a hardware store are ideal.  When the sealant has hardened sufficiently, sealant extending beyond the edge of the cover glass can be carefully removed if desired using a razor blade.  (12) As the sealant hardens, some inwelling may occur at the margins of the cover glass; airspaces thus formed can be filled with sealant or with nail lacquer.  (13) The slides are labeled and stored flat.

Cultivation.  As was mentioned earlier, no member of the Laboulbeniales has been cultivated from spore to spore despite several attempts. There is, however, one report of cultivation of Stigmatomyces ceratophorus (Whisler 1968).  Autoclaved wings of the host, the lesser house fly, Fannia canicularis, placed on brain-heart infusion agar fortified with typtose, a mixture of enzymatically digested protein, and overlain with horse serum were used to grow thalli up to a 20-celled antheridium-producing stage.
 Preparation and deposition of vouchers.  Well-sealed slide mounts should be deposited in collections.  Data should be written legibly in permanent ink on a label securely attached to the slide.  Host information should be included so that voucher material of the host can be located years later if necessary.  Vouchers of host specimens should be kept in 70% alcohol, preferably with the slide collection or linked to it by a common collection number.  Addition of some glycerol to the alcohol will retard evaporation of the solution; it is imperative, however, to have containers that are well sealed.  Plastic vial with caps having o-rings are especially suitable for long-term storage.
     The Farlow Herbarium at Harvard University is the repository for the extensive R. Thaxter collection.  Other large collections made by R. K. Benjamin and I. I. Tavares are in the herbaria of the Rancho Santa Ana Botanic Garden, Claremont, California, and University of California, Berkeley, California, respectively.

Identification.  Only very few specialists attempt to identify members of the Laboulbeniales to species, and it is difficult to know when one has collected a new species.  Many keys are based on the system of Tavares (1985), and host identification is an important aid in the identification process.  Tavares (1985) recognized 132 genera: one in the Herpomycetaceae, 12 in the Ceratomycetaceae, five in the Euceratomycetaceae, and the other 114 in the Laboulbeniaceae.  Since the publication of Tavares's book, six additional genera [i.e., Majewskia Lee and Sugiyama (1986), Sugiyamaemyces I. I. Tav. and Balazuc (1989), Cupulomyces R. K. Benj. (1992a), Phalacrichomyces R. K. Benj. (1992b), Corylophomyces R. K. Benj. (1995), and Parvomyces Santam. (1996)], all Laboulbeniaceae, have been described and one genus, Fanniomyces has been reduced to synonymy under Stigmatomyces (Weir and Rossi 1995) for a total of 137.  Good evidence from ongoing collection studies (Rossi and Weir 1997) and from quantitative assessment of well-inventoried tropical beetle faunas (Weir and Hammond 1997a,b) indicates that large numbers of undescribed species exist.
     Tavares (1985) recognized two suborders of Laboulbeniales based on the origin of the centrum in relation to the development of the perithecium wall.  In the Herpomycetineae, with only one family, Herpomycetaceae, the perithecium wall begins to develop prior to the appearance of the carpogonial upgrowth, whereas in the Laboulbeniineae, which includes the other three families, Ceratomycetaceae, Euceratomycetaceae, and Laboulbeniaceae, the carpogonial upgrowth is formed prior to the appearance of the first perithecium wall cells.  The latter three families are distinguished from each other by the relationship of their perithecial stalk cells to the cells of the receptacle during early stages of development of the perithecium.  Genera are distinguished by differences in the morphologies of their receptacles, appendages, antheridia, and perithecia, as well as the positions of their appendages, antheridia, and perithecia on the receptacle.  The literature for identification begins with the monographic studies of Thaxter (1896, 1908, 1924, 1926, 1931) in which many species are beautifully illustrated.  The most recent complete key to genera can be found in Tavares (1985), who discusses the genera and provides numerous references to primary literature.

Fungi of unknown affinity
A number of fungi found on the surfaces of arthropods that do not appear to harm.  The fungi are observed only by specialists interested in them specifically or by those studying Laboulbeniales.  In fact many of these species have been seen by few mycologists since R. Thaxter and C. Spegazzini described them more than 50 years ago (see Blackwell and Rossi 1986; Blackwell et al. 1986; Blackwell 1994).
    Genera of this group appear to be specialized on termites (Termitaria, Mattirolella, Laboulbeniopsis, Coreomycetopsis, Antennopsis, Hormiscioideus), attine ants (Termitariopsis), and other insect hosts (Muaiaria, Muriogone, and Chantransiopsis).  Little is known of their biology.  Some, but not all, have haustoria that penetrate the host cuticle.  The few characters available for these morphologically reduced forms indicate that they do not form a monophyletic group.  A few of them may have alternate states or growth forms that should be sought in the habitats of the host insects (James W. Kimbrough pers. comm.).  This is the situation with Thaxteriola, an ascospore-derived anamorph to be discussed later in the chapter (Spore dispersal interactions).
      Because these species have much the same habit as the laboulbenialean species discussed above, the techniques for their study are essentially similar.  However, in many cases the fungi are much smaller than Laboulbeniales.  A high quality dissecting microscope capable of magnifications of at least 200X and equipped with fiber optics makes discovery and mounting of these forms easier.

Septobasidiales
A number of ascomycetes are serious pathogens of scale insects (Homoptera: Coccidae); species of Septobasidium, however, appear to have a mutualistic relationship with their insect associates and to cause them little harm.  A colony of scale insects can be entirely covered by the fungal colony and thus derive partial protection from predators.  In return a few individuals of the colony are invaded by helical haustoria of the fungus.  Infected individuals do not die; in fact they may outlive uninfected neighbors.  The only apparent disadvantage is that they cannot reproduce.

Collection.  Species of Ordonia, Septobasidium, and associated anamorphs (Janetia and Johncouchiaspp.) form perennial colonies on the surfaces of plant structures with colonies of scale insects.  The basidiocarps are usually brown to black, rarely more brightly colored, normally resupinate, and felty in texture.  Their surfaces may be smooth, warty, or spiny.  Inexperienced collectors may mistake them for corticioid basidiomycetes or even lichens.  Septobasidium occurs on living leaves, stems, and branches of a great variety of perennial plants, including gymnosperms, monocots, and dicots.  As is the case with most basidiomycetes, basidia are produced so that they project toward the ground.  Thus the resupinate basidiocarps often are found on the lower sides of branches.  Their occurrence on living rather than dead plant parts, and away from the extreme tips of branches distinguishes them from some other resupinate species of Aphyllophorales.  Coccidiodictyon inconspicuum differs from other members of Septobasidiaceae in producing an inconspicuous colony that is not easily seen in the field and probably will be found more readily by examining collections of scale insects with a dissecting microscope.
    Specimens of Septobasidiaceae can be collected using methods suitable for collecting Aphyllophorales or lichens; that is, they are normally removed from the substrate with a knife or ax, and should include a generous portion of substrate and associated scale insects.  Because the identity of the substrate plant is essential for identifying the associated scale insect and the fungus, a voucher specimen of the plant should be collectedand include a sample of leaves, flowers, fruit, and other diagnostic structures.  The materials should be placed in paper bags or in well-aerated containers, not plastic bags.
     Septobasidiales are perennial and thus, exhibit distinct seasonal responses.  Growth occurs during the wet season and ceases or slows at the onset of the dry or cold season.  Probasidia (telia) are produced at the end of the growing season or during the dormant period.  Germination of probasidia occurs at the onset of the new growing season.  Because basidia are taxonomically important structures, the most useful material for identification will be collected toward the end of the dormant period or at the beginning of the growing season.

Storage.  Most Septobasidiales tolerate dryness.  Their perennial colonies are adapted to periodic drying, and probasidia function as dormant perennating structures.  Nevertheless, drying over extreme heat will kill a probasidium and prevent subsequent development.  Air drying is preferable in most cases, although in very humid climates it may be necessary to use desiccants or even low heat.  The drying air should be moving, not stationary as it is in an oven.  The top rack of a large mushroom dryer is usually sufficient in any climate.
    Identification of the fungus may require identification of the host insect as well as examination of haustoria, probasidia, basidia, and basidiospores. Separating a colony from the insects requires delicacy, and it is best to soak the material in water first to thoroughly wet it.  Couch (1938) recommended pulling a colony away from the insects with the specimen still submerged in water.  In this way only the connections between the colony and parasitized insect individuals will remain.  Finding parasitized individuals, which may be greatly outnumbered by healthy ones, is important if one is to observe the taxonomically diagnostic haustoria.  Parasitized individuals may be smaller than healthy ones, but not necessarily so.
     Healthy scale insects should be saved for identification although only adult females can be identified reliably. The insects are dried with a bit of the plant substrate or cam be placed in alcohol.

Preparation of specimens for study.  The critical structures used for identification can be observed by mounting. Haustoria were divided by Couch (1938) into six types.  These are important taxonomically and must be observed carefully.  In most cases whole infected individual insects can be mounted on a microscope slide in water; they are transparent enough to allow internal structures to be viewed.  If the insects are dry, they can be mounted in 7% KOH, or if they are very old and dry, even boiled in KOH.  In some cases it may be necessary to crush the insect or even dissect the exoskeleton away to reveal the haustoria.  The port of entry (or exit) of the fungus into the insect also may be important and should be noted.
     Probasidia, basidia, and basidiospores are critical in identifying Septobasidiales.  Basidia may be 1-, 2-, 3- or 4-celled, and the probasidium may be persistent or not.  Basidia may be straight, bent, or coiled.  Mature basidiospores are actively discharged from the sterigmata.  Unless the material has been collected at precisely the right moment it will not exhibit basidia.  However, many live specimens can be induced to produce basidia and basidiospores.  The specimens are soaked in water until they are well wetted and then placed in a covered container with wet paper towels.  Within 24 to 48 hours the probasidia may undergo development and produce basidia and basidiospores.  These can be mounted in water or 7% KOH for examination.

Cultivation and deposition of cultures.  Cultures can be established by moistening a colony producing probasidia and suspending it over low nutrient or water agar.  This is done easily by attaching a small fertile portion of the basidiocarp to the top of a petri dish.  Some workers use white glue for this purpose, but a piece of agar cut from the edge of the plate often adheres to the lid and can be used to hold the basidiocarp.  This procedure requires less care than one might imagine, but crumbs from the specimen that carry faster-growing fungi must not be allowed to fall on the agar surface where the ejected basidiospores are expected to germinate.  Common laboratory media such as malt-yeast-peptone used to culture many "jelly fungi" appear to be adequate.  As with any group of fungi, cultures should be deposited in an established culture collection.  In view of the ease with which at least some species can be cultivated it is rather surprising that few of these fungi are available in culture collections.

Preparation and deposition of vouchers.  Vouchers should consist of all or part of a colony, including the associated insects and some of the substrate.  These can be placed in boxes or packets of the kind used for lichens and Aphyllophorales.  Labels should include the names of the fungus, the associated insect, and the plant substrate, as well as the date, and usual geographical and habitat data.  The J. N. Couch collection is at the University of North Carolina; however, until there is an active mycologist at that institution, we cannot recommend sending vouchers there.  A more appropriate collection that contains a large collection of Septobasidium is the Farlow Herbarium.

Identification.  Characters used in identification are discussed briefly above under the section  “Preparation of specimens for study.”  Unfortunately, there is remarkably little literature on Septobasidiales.  Couch's 1938 monograph is still the most authoritative source, and few species have been described since that work was completed. However, Couch recorded the largest number of species (36) from the United States, in spite of the fact that these fungi probably are predominantly tropical.  Collectors in tropical regions likely will discover a large number of new taxa, a point Couch (1938:50) recognized when he stated that "no discussion of geographical distribution will be of much value." Other useful references to Septobasidiales are those of Couch (1935) and Azema (1975).

Mutualistic associations between insects and the fungi upon which they feed or from which they acquire enzymes for digestion, are often referred to as gardening symbioses (Martin 1987). We also should point out, however, that not all of the fungi in these associations are strictly members of gardening symbioses, but rather there is a continuum of associations ranging from dispersal to true gardening associations. Some of the associations are of interest because they may provide systems for evolutionary studies of a spectrum of interactions. In other cases the interactions are of economic importance because they involve dispersal of serious fungal pathogens or sapstain fungi that damage trees, crop plants, and forest products.  Some of these fungi rely on the insect for survival, because they are poor competitors with saprobes in their habitats. The fungi include ascomycetes that are symbionts of various groups of beetles (yeasts, Ophiostoma, Ceratocystis and related conidial forms, and aphyllophoralean basidiomycetes), intracellular yeastlike forms that inhabit specialized host cells and coelomic cavities (Symbiotaphrina and undescribed taxa), and the basidiomycetes and ascomycetes associated with siricid wood wasps, ants, and termites (Table 4).

Table 4.  Fungi involved in gardening symbioses with arthropods. Ascomycota Saccharomycetes Saccharomycetales ( Ascoidea, Dipodascus, Pichia, Candida) Pyrenomycetes Hypocreales (undescribed yeastlike forms associated with planthoppers) Xylariales Xylariaceae (Xylaria) Microascales Ceratocystiaceae (Ceratocystis, Chalara, Ambrosiella, in part) Ophiostomatales Ophiostomataceae (Ophiostoma, Leptographium, Ambrosiella, in part, Sporothrix, Raffaelea) Loculoascomycetes-Discomycetes Unknown affinites - Symbiotaphrina Basidiomycota Hymenomycetes Aphyllophorales Corticiaceae (Entomocorticium and others) Agaricales Lepiotaceae (Chlorophyllum, Leucoagaricus, Termitomyces and undescribed forms)

     In this section we consider fungi associated with the beetles that inhabit bark and wood of living or recently dead trees. We include the fungi that commonly occur with phloem-feeding beetles, usually in living trees (grouped here as bark beetles), and fungi associated with beetles that require a fungal primary nutrient resource in all life history stages (ambrosia beetles).  The distinction is artificial, and the fungi often are closely related.  We also discuss phloem-feeding weevils (Curculionidae) with the bark beetles.  Because techniques used to study the phloem-feeding and ambrosial associations differ somewhat, we will discuss them separatly.

Bark beetles and fungi
Bark beetles colonize both hardwood and conifer trees, and although we will emphasize those that colonize conifers and their fungal associates, many of the methods we describe are applicable to colonizers of hardwoods as well. Numerous fungi may be found occupying almost all parts of the body surface and gut of a beetle, as well as the tree tissue the beetle infests.  Among the fungi found on the beetle surface and within the digestive tract are yeasts (Calaham and Shifrine 1960; Bridges et al. 1984; Leufven and Nehls 1986), various saprobes (Bridges et al. 1984), and ophiostomatoid fungi (see section on identification below), especially in Ophiostomatales (Upadhyay 1993). Species of Ophiostoma and related conidial fungi associated with beetles include many of the stain fungi known to discolor wood (Harrington 1988 and references therein).  Beetle-associated ophiostomatalean fungi also have been implicated as conifer pathogens (Harrington and Cobb 1988; Harrington 1993), and certain members of this group are capable of killing trees (Brasier 1988; Harrington 1993; Solheim et al. 1993).  More often, however, they are associated with resinous lesions that may cause the occlusion of sapwood (Harrington 1993).  Some of these fungi also are antagonists of beetles, reducing reproductive success and larval development (Barras 1970).  Although the exact roles played by the various ophiostomatoid fungi have yet to be determined, they are undoubtedly closely associated with bark beetles and weevils and with their tree hosts.  Efforts to examine the diversity of beetle-associated microorganisms centers around these fungi.
     Certain beetles of the families Scolytidae and Platypodidae have evolved specialized structures known as mycangia, the purpose of which appears to be the storage, cultivation, and transport of fungi (mycangia occur also in ambrosia beetles). The mycangia of a few bark beetle species are complex and include secretory cells (Harrington 1993). More commonly, beetle mycangia are less well-developed, simple pits in the exoskeleton of the head, pronotum, or elytra.  These simple structures may contain yeasts, ophiostomatalean fungi, and other fungi, including corticioid basidiomycetes (Harrington 1993; Lewinsohn et al. 1994). Mycangial fungi are thought to be mutualists of their beetle hosts, possibly by having nutrients provided by the host (Bridges 1983; Bridges and Perry 1985; Goldhammer et al. 1990). Often fungi have a yeast-like morphology while they are in the mycangia, rather than hyphal form outside the mycangia and in the environment of the wood. The taxonomy and ecology of many of these fungi are not fully known (Moser et al. 1995).

Collection.  The bark beetle fungi may be found in or on insects,other than the beetles that they colonize.  Effective collection of specimens for isolation of fungi requires knowledge of the host insect's biology.  Hosts of these fungi are found in only a few coleopteran families: Scolytidae (bark beetles), Curculionidae (weevils), and Platypodidae (Harrington 1988; Malloch and Blackwell 1993).  These insects may colonize the lateral roots, the root collar, the main stem, the branches, developing shoots, and even fruits of a variety of trees (S. L. Wood 1982; Drooz 1985); the best-studied insect-fungal complexes are found in conifers.  Many bark beetles and weevils use tree- and insect-produced compounds to locate suitable hosts as well as mates (D. L. Wood 1982). Using host material, host compounds, and/or beetle pheromones during times of seasonal insect abundance, it is possible to collect large quantities of beetles from which fungal associates may be isolated. Adult beetles actively seeking host substrate and/or mates respond to indicators of attacking beetles and/or susceptible trees. Aggregation pheromones, either alone or in combination with host compounds, serve as potent attractants for such beetles (D. L. Wood 1982).  Later arriving other beetles and weevils (secondary invaders), ordinarily associated with stressed or dead trees, may respond strongly to host volatiles (Tunset et al. 1993).  A variety of sampling devices has been designed to capitalize on these key aspects of bark beetle and weevil biology.
     Pitfall traps have been used to capture root-infesting weevils and beetles within forests and plantations (Harrington et al. 1985; Witcosky et al. 1987; Hunt and Raffa 1989).  One such trap (Hunt and Raffa 1989) can be constructed from capped sections of plastic pipe, drilled with small entrance holes and placed so that the holes are even with the soil surface.  Vials of ethanol and turpentine are hung inside the trap, and a section of pine stem is placed in the bottom of the trap.  Weevils and beetles crawl through the soil and into the entrance holes in response to the volatiles (Hunt and Raffa 1989) and are unable to escape.  Stem sections (billets) also have been used to collect root-feeding beetles and weevils (Lewis and Alexander 1986; Tunset et al. 1988).  The stem section usually is placed in contact with the soil where it is left for several days.  The host material is checked daily for insects moving from the soil onto the stem section surface.  In addition, some root weevils may be collected from the lower stem as they ascend at night to feed on branches (Klepzig et al. 1991).  Walking weevils are forced into a collection jar atop a screen funnel wrapped around the main stem.
     Bark beetles that attack the lower stem of trees may be collected in various types of flight traps.  A lower stem flight trap consisting of an inverted, plastic jug modified by having a collection jar attached may be baited with turpentine and ethanol and used to collect turpentine beetles and some root insects (Klepzig et al. 1991).  Turpentine beetles also may be captured in bounce traps in which the flying beetle strikes a black pipe baited with ethanol and turpentine and is collected from a water-filled pool below (Fatzinger 1985; Phillips et al. 1988).  Lindgren multiple funnel traps (Lindgren 1983) can be hung near ground level for collection of lower-stem insects flying toward attractants (Phillips et al. 1988).
     Most of the aggressive, tree-killing bark beetles attack the central and upper portions of the stem (S. L. Wood 1982; Drooz 1985).  Flight traps hung in the mid to upper canopy and baited with species-specific pheromones often are employed to sample these insects.  The proper choice and use of pheromones is, however, a complicated matter.  Species may respond to different compounds or to the enantiomers of those compounds (S. L. Wood 1982; Payne et al. 1984; Raffa and Klepzig 1989).  In addition, methods used to sample populations of the beetles often are at odds with obtaining viable cultures of associated fungi.  As an example, collection jars of Lindgren multiple funnel traps can be filled with soapy water (Klepzig et al. 1991) or an insecticidal strip (Hayes and Strom 1994) to kill trapped beetles.  Both of these methods are likely to influence the fungal flora of the beetles.  In addition dead beetles soon become overwhelmed by saprobic fungi that may interfere with isolation of beetle-associated fungi.  An alternative is to leave the collection jar empty so that beetles are not killed upon falling into the trap.  However, bark beetle pheromones may serve as kairomonal attractants for predaceous insects, in which case the collection jars become a feeding ground for predators (S. L. Wood 1982).  The result is a pile of bark beetle parts with little value as a source of fungi. If traps are sampled the same day they are deployed the method may, however, be used successfully. Alternately, daily collection from a collection cup contained within an electric cooler to slow the predators has been recommended (B. S. Lindgren pers. comm. 1996).
     Many of the bark beetle-associated fungi may be isolated from infested host tree material.  Phloem and xylem tissue from areas around insect feeding sites, entrance holes, and adult or larval galleries may be collected aseptically for fungal isolation as can frass from larval galleries (Bridges et al. 1984; Harrington 1992; Solheim 1995).
     Various phoretic tarsonemid mites have been implicated in the transmission of bark beetle-associated fungi, and these too may be a source of cultures (Bridges and Moser 1983; Moser et al. 1995).  Mites are typically removed from the beetle exoskeleton (often underneath the beetle elytra) with fine needles.  They can be cultured directly or mounted on slides for later microscopic examination of spores.

Storage. Isolations should be conducted as soon after the insects are collected as possible.  When this is not feasible, insects, mites, and tree tissue should be placed in sterile vials, and transported to the laboratory in ice-filled coolers, and refrigerated until the fungi are isolated are possible.  Placing small pine twigs and/or moist paper in collection vials may increase survival of insects during periods of extended refrigeration.  When possible, insects should be kept in separate vials or containers to minimize the likelihood of cross contamination between hosts.  Tree-tissue samples can be refrigerated or stored at room temperature in moist chambers, thus helping to keep the fungi viable and actually may promote sporulation of the fungi on the host substrate (Seifert et al. 1993).  Specimens can be dried, but subsequent isolation into culture often is not successful.

Preparation of specimens for study.  Diagnostic characters for identification of the fungi are obtained from structures associated with sexual or asexual sporulation in the tree wood or bark.  It is important to make good slides of material when it is available even if it is to be cultured, because some of these species will not produce sexual reproductive structures in culture.  To serve as vouchers, specimens must be mounted on microscope slides using the double cover slip method discussed above (see "Preparation of specimens for study" in the Laboulbeniales section for detailed protocol), and made permanent.  In addition, spores of fungi found in pits or mycangia on beetle and mite exoskeletons have been examined and in some cases identified using light and transmission and scanning electron microscopy (Happ et al. 1971; Lewinsohn et al. 1994).

Cultivation and deposition of cultures.  Mycangial fungi may be isolated from beetles by dissecting the mycangia from the beetle, surface sterilizing the structure, and subdividing and placing it on selective media (Barras and Perry 1972).  For example, for the southern pine beetle, the pronotum is removed from the head and abdomen of an adult female and the legs are removed from the pronotum using a pair of sterile, fine forceps.  The entire, intact pronotum is placed in sterile distilled water for one minute, placed in modified White’s solution (1.0 g HgCl2, 6.5 g. NaCl, 1.25 ml HCl, 250 ml 95% ethanol, and 750 ml sterile distilled water; Barras 1972) for four minutes, and then passed through two successive rinses with sterile distilled water.  The pronotum then is quartered aseptically with fine forceps, and the four sections are placed on malt extract agar (MEA) or MEA amended with 2 mg/ml benomyl to facilitate the semiselective isolation of certain species such as C. ranaculosus, an isolate known as SJB 122, and perhaps other mycangial fungi (Ross et al. 1992).
     Beetles can be crushed on an agar medium to isolate phoretic fungi (Harrington 1992).  Alternately, beetles may be ground in sterile distilled water in a glass tissue homogenizer.  Aliquots of this homogenate are plated directly or diluted a number of times and then plated on an agar medium (Klepzig et al. 1991).  This technique has the added advantage of allowing for quantitative estimates of the number of propagules of each fungal species being carried by each insect.  Samples of beetle-infested tree tissue may be aseptically collected and placed directly on water agar (WA).  In many cases, fungal fruiting structures may be found lining beetle galleries, and a fine needle may be used to transfer masses of spores directly to media (Seifert et al. 1993).
     A major confounding factor in all of the isolation techniques described is the presence of saprobic contaminants either on the beetle exoskeleton or within beetle-infested tree tissue.  Although these fungi may be significant components of the beetle fungal flora, many of them grow so quickly that they overwhelm other fungi of interest (Gibbs 1993; Seifert et al. 1993). Diluting samples before plating often results in separation of fungal colonies of interest can by subcultured (Klepzig et al. 1991). Another technique useful for isolating members of the Ophiostomatales, which includes many of the well studied beetle associates.  A characteristic of this group of fungi is a high degree of tolerance of the antifungal compound cycloheximide (Seifert et al. 1993).  Accordingly, malt extract, WA, or potato dextrose agar (PDA) may be amended with 200 ppm cycloheximide, to inhibit growth of nonophiostomatalean fungi, and 100 ppm streptomycin sulfate to inhibit bacterial growth (Harrington 1992; Seifert et al. 1993). Some yeasts, however, and species of filamentous fungi, including Penicillium, also may grow on these media (Harrington 1992).  Single colonies on plates may be transferred (via hyphal tip transfer and/ or transfer of spore masses) to unamended PDA or MEA for identification of pure cultures.
 Plates are normally incubated at from 20 to 25°C, although some bark beetle associates may grow better at cooler temperatures (Harrington 1992).  Cultures may be grown on a weak agar medium in vials and stored at -20°C, or they may be lyophilized for preservation.  Serial transfer and storage on rich media are not generally recommended (Seifert et al. 1993).
     All of these fungi would be of interest to most general culture collections because of their interesting biological associations and economic importance.

Preparation and deposition of vouchers.  Vouchers can be prepared by placing fungal structures on the plant material or cultures that have been dried, usually in moving air at room temperature, in packets or small boxes.  The specimens should have sexual reproductive structures that include mature diagnostic characters.  It also is desirable to prepare permanent slide mounts as vouchers.  In some cases parts of the beetles and associated mites can be mounted or prepared for scanning electron microscopy to show phoretic spores of the fungi.  Arthropod material can be maintained in about 70% alcohol to which a little glycerine has been added.

Identification.  The fungi that occur in bark beetle associations are a diverse lot.  We mentioned ophiostomatoid fungi with morphological features such as evanescent asci and long-necked ophiostomatoid perithecia through which the sticky ascospores are passively discharged for arthropod dispersal.  These species belong to several orders, primarily Ophiostomatales (Ophiostoma), but also Microascales (Ceratocystis), and Pyxidiophora.  An unusual fruiting structure produced by Heterogastridium, a basidiomycete, can be mistaken for one of the ophiostomatoid ascomycetes at first glance (T. J. Perry pers. comm.).  In addition among these orders there are a number of derived asexual forms such as Sporothrix, Leptographium, and Chalara.  There are yeasts, mycangial basidiomycetes, and other filamentous saprobic fungi as well.  Several helpful references to these fungi are available (Barras and Perry 1975; Harrington and Cobb 1988; Perry 1991; Schowalter and Filip 1993; Upadhyay 1981; Wingfield et al. 1982; S. L. Wood 1982).

Ambrosia beetles and fungi
Wood-boring scolytid and platypodid ambrosia beetles usually inhabit dead or dying trees [see reviews by Batra (1967), Francke-Grossmann (1967), Norris (1979), Beaver (1989), and Roeper (1995)].  Many of the fungal genera that we discussed in the section above on bark beetles and fungi are associated with the same families of beetles in these stricter interactions.  As was mentioned earlier, the primary difference between the ambrosial habit and that of the bark beetle habit is one of nutrient resource with adult, pupal, and larval stages of ambrosia beetles relying on fungi as the primary food resource.  Roeper (1995) described two feeding categories of ambrosia beetle larvae: those that consume only fungi (mycetophagous) and those that enlarge their gallery or larval cradles in the xylem as they develop, thus consuming both wood and fungal material (xylomycetophagous).  This approaches the behavior of some mycangial bark beetles.  As adult beetles make brood galleries by tunneling into the new woody host material, they transmit species-specific obligatory fungal symbionts in ectodermal mycangia.  The damage done by ambrosia beetles is from the boring activity and subsequent fungal staining (McLean 1985).
     Primary symbionts of a beetle are those that are consistently isolated from the mycangia of adult beetles collected during flight, from adults excavating their new brood galleries, or from brood galleries in the presence of actively feeding larvae (Batra 1967, 1985; Roeper et al. 1980).  These fungi include Ambrosiella and Raffaelea, asexual genera related to Ophiostoma and Ceratocystis.  In warmer regions Fusarium may be an ambrosial associate.  Other fungi referred to as auxiliary or secondary ambrosial fungi are not usually isolated from mycangia but are regularly present in the brood tunnels only after beetle pupation.  Many of these fungi have mucilaginous spores that may be transmitted by phoretic mites and beetles; their presence parallels that of the saprobic fungi associated with bark beetles and interferes with the isolation of the slower-growing primary symbionts in culture.
     S. L. Wood (1982) described the North and Central American Scolytidae and listed their woody hosts and geographical distributions.  In addition, Wood and Bright (1987, 1992) have catalogued Scolytidae and Platypodidae and their plant hosts, distribution records, and literature references world-wide.

Collection.  Ambrosia fungi are collected with their beetle associates.  The fungal form is dimorphic; in the well-developed mycangia the ambrosia fungus is yeastlike and within the galleries it is filamentous, but usually cropped by the feeding beetles.
    The beetles can be collected during dispersal flight, and in northern temperate regions, adult beetles fly only during a short time period (generally a month) in spring when temperatures reach 18C (Roling and Kearby 1975; Turnbow and Franklin 1980; Weber and McPherson 1991).  Most have only a single annual flight, but some species (e.g. Monarthrum spp.) have two generations and two flights each year.  In semitropical and tropical regions, many of the species have multiple generations, but seldom fly during dry seasons.  The beetles tend to fly in the late afternoon and early evening.
     Beetles can be caught live in fine mesh nets, in mechanical rotary traps (Rudinsky and Daterman 1964), or with hand nets.  Unprocessed timber at sawmills, logging operations, wind-thrown, wind-damaged, standing suppressed and/or diseased trees attract flights of ambrosia beetles.  The beetles are attracted to ethanol produced by the fermentation of host timber.  Thus ethanol or beer can be used in addition to timber to attract beetles to a collection site. Beetles collected should be placed individually into sterile vials or stoppered tubes with damp sterile filter paper and then cooled during transport to the laboratory.
    Most ambrosia beetles in temperate regions infest cut timber, wind-thrown trees, wind-broken limbs or boles of trees, and/or suppressed or diseased trees.  Woody timber dead more than a year is seldom infested.  However, ambrosia beetles of the genus Corthylus in temperate regions and members of the scolytid beetle tribe Xyleborini (Xyleborus and Xylosandrus) in warm temperate, semitropical, and tropical regions are capable of attacking apparently healthy and undamaged woody hosts.  Ambrosia beetles infest the bole of the timber and bore directly through the bark into the xylem.  The entrance hole is seldom more than 1 mm in diameter; the boring frass is light in color initially, but darkens as the primary symbionts begin to grow or it is contaminated with larval fecal pellets.  By comparison, scolytid bark beetles typically produce brownish boring frass because they mine the inner bark as they construct galleries.

Storage.  Once infested timber is located, it should be cut into manageable lengths, generally to about a meter, and returned to the laboratory.  The infested timber is cleaned of surface dirt and biota, and disinfected by wiping the surface lightly several times with ethanol.  Once the beetles have bored into the xylem they generally will not reemerge unless the wood begins to dry out.  Painting cut ends with melted paraffin wax slows log dehydration.  If stored out of direct sunlight, the beetles will continue gallery construction, produce broods, and complete a generation of their life cycle.

Preparation of specimens for study.  These methods generally are the same ones used for bark beetle associates; however, diagnostic characters such as sexual stages and even conidia may be lacking.  Cultures are important for identification of these species and provide additional characters such as growth rate and pigment production.
     Isolation of symbiotic fungi from galleries.  The primary fungi of an ambrosia beetle are abundant in the gallery only when larval stages are present (Kajumura and Hijli 1992).  Thus, the best isolates of primary fungal symbionts can be made a month or two after initial infestation.  Galleries are exposed by sawing thin sections from the infested bole.  It is important to work as quickly and as aseptically as possible, using alcohol flamed saws, wood chisels, and/or pruning shears.  Adult insects can be removed and isolations made from visible fungal growth within the several mm diameter gallery using sterile fine forceps.  Samples of thin slices or chips of galleries should be preserved, dried and/or mounted directly on slides with fixative mounting medium such as lactophenol-aniline blue for later study.
     Ambrosia fungi from Corthylus and most Xyleborus species generally form a thick whitish palisade layer of growth on the walls of galleries if eggs and/or larvae are present.  This fungal growth can be isolated easily by streaking or spot plating on isolation media (see following).
Fungal growth usually is not so evident on the gallery walls or larval cradles, of xylomycetophagous insects; thus, small slices and chips of wood should be aseptically removed for plating.  Slices or fragments of galleries can be placed aseptically in a sterile moist chamber to encourage fungal growth in the absence of actively feeding larvae, so that primary ambrosia fungi can be isolated, often within a few days, before contamination from saprobic fungi.
     Live beetles from flight and/or galleries are difficult to handle because of their small size and smooth cylindrical shape.  A vacuum simple apparatus consisting of a sterile micropipette tip with small aperture attached to a rubber hose fixed to a vacuum pump or vacuum line allows one to pick up individual beetles and transfer them easily from dish to dish or to sterile glass slides for dissection.
     Beetles can be surface disinfected by washing in sterile 0.1% HgCl2 solution or dilute sterile bleach (NaHCl2) for 2 to 4 minutes to reduce nonmycangial microbes, followed by several rinses in sterile water.  Adult beetles also can be freed of external nonmycangial microbes by placing them alternately in plates of sterile wet filter paper for 18 hours and then on dry sterile filter paper for 6 hours.  Several transfers typically will free them of most external microbes.  Individual beetles can be stored on sterile moist filter plates for months at refrigerator temperature until needed for dissection and isolation.  Prevention of dehydration appears to be the critical factor for their long-term storage.

Culture.  Primary ambrosia fungi are found abundantly in the mycangia at the time of flight and/or early stages of gallery development (Roeper 1988; Kajimura and Hijli 1992).  For this reason timing is important for isolation of the true primary symbiotic fungi from a beetle.  Beetle sex also is important because mycangia usually develop only in the sex that initiates the brood gallery system.  The beetles usually have a single pair of mycangia, and their position can vary between even very closely related beetles. Oral, pronotal, mesonotal, prothoracic pleural, pro-mesonotal, or elytral mycangia of different scolytids and the pitlike mycangia of platypods should be dissected from the beetle and the contents plated onto agar medium for isolation.  Beetle dissection should be done on sterile alcohol-flamed glass slides in 3 separate drops of sterile saline or bovine serum using alcohol-flamed fine watch-maker forceps, fine needles, and sterile micropipettes under a dissecting microscope.  The body part of the beetle containing the mycangium(a) should be separated from the other body parts in the first of the three drops.  In the second drop, the mycangium(a) then can be separated.  After transfer to the last drop, the mycangium(a) can be broken apart and plated. The presence of fungal cells can be verified using the low power of a compound microscope.  Mycangial fungal cells sink to the bottom of the drop, whereas insect fat droplets with which they might be confused, stay in suspension.  The mycangial fungi tend to be yeastlike budding forms and/or monilioid chains if the fungi are proliferating actively in the mycangium.  All parts of the mycangia are then spot plated and/or streaked on isolation media (see following) and incubated at 22 to 25°C.  Sterile micropipettes can be used to pick up small masses of fungal material to be plated.
     If the location of the mycangia is unknown, then careful microscopic examination of all the beetle's body-parts is required.  Whenever fungal cells are discovered additional sterile slides and drops of dissection medium are needed to separate fungus from the mycangia of the dissected beetles.
     Precoxal mycangia (e.g., in species of Monarthrumand Gnathotrichus) can be dissected to remove their mycangia or they can be sampled directly. The adult beetle is killed and fixed ventral side up on a sterile glass slide with a drop of molten paraffin.  Under a dissecting microscope the forelegs are removed and then a sharp needle is used pick out the contents of the enlarged coxal mycangium; this material is plated directly onto isolation medium.
    Examination of plates should be done daily after isolations have been made from a gallery or mycangium.  Hyphal-tip subisolation of filamentous fungal growth and restreaking of yeastlike colonies usually is necessary for purification.  Some Ambrosiella species (A. hartigii, A. ferruginea, A. xylebori, A. sulphurea) grow rapidly in filamentous form and often produce melanin pigments.  Ambrosiella brunnea, A. gnathotrichi, and most Raffaelea species form yeastlike mycelial colonies initially and should be subcultured by streaking and hyphal tipping.  Yeasts are commonly encountered and should be subcultured by streaking.
    Several culture media can be used to culture primary mycangial symbionts: potato dextrose agar, yeast extract-malt extract agar (30 g malt extract, 10 g yeast extract/liter of distilled water), or dilute yeast extract-malt extract-glucose agar (10 g malt extract, 5 g yeast extract, and 5 g glucose/ liter of distilled water) .  As many isolations as possible should be attempted from available collected material.  Frequency of occurrence of a particular microbe should establish the presence of associated symbiont microbes.  Prokaryotic microbes are seldom encountered, so antibiotics generally are not used in isolation media.  Once axenic cultures have been made, they can be stored on slants of dilute yeast extract-malt extract-glucose agar for future study.

Preparation and deposition of vouchers.  Procedures for preparation and deposition of vouchers are similar to those for bark beetle associates; however, sexual stages are not produced in these fungi.  It is important to prepare vouchers from early cultures, because ambrosia fungi may stop producing conidia after a few transfers.

Identification.  Primary fungi (Ambrosiella and Raffaelea species) can be identified using the works of Batra (1967) and Roeper et al. (1980).  Identification of the beetle associate is important because the fungi usually show host specificity.  Many of the filamentous ambrosia fungi fail to sporulate in culture or after repeated subculture.  However, increasing nitrogen content (L-proline) of the medium, buffering the agar to a pH near neutral, and elevating CO2 levels in the of the culture may induce sporulation in the fungi. Molecular studies have been used to characterize species of Ambrosiella and Raffaelea and have shown that Ambrosiella is polyphyletic with some species being related to Ceratocystis and others allied with Ophiostoma (Cassar and Blackwell 1996). A similar study has shown that Raffaelea species are related to Ophiostoma (Jones and Blackwell 1998).  Reference cultures of most known primary ambrosia fungi are available from ATCC and CBS.

Yeast-like Endosymbionts and Extracellular Symbionts
The term endosymbiotic is used to denote the situation in which one organism, here an insect, harbors an intracellular symbiont, here a fungus, usually within cells (mycetocytes) of specialized structures (mycetomes); in some cases mycetomes are modified fat bodies.  The yeast-like fungi are single-celled and reproduce only asexually.  Yeast-like symbionts are associated with anobiid and cerambycid (long-horned) beetles (Jurzitza 1979; Nardon and Grenier 1989), planthoppers, and some species of scale insects.  The symbionts of cerambycid beetles and of some anobiid beetles have been cultured.  Extracellular hemocoel-inhabiting symbionts have been observed in a group of gall-forming social aphids (Fukatsu and Ishikawa 1992) and in a wasp (Comperia merceti) that parasitizes cockroaches (Lebeck 1989; Table 5).  Although these fungi have garnered little attention from mycologists, they have a fascinating biology.  The associations apparently have arisen independently from several fungal lineages.

Table 5.  Primary taxonomic groups of insects that harbor yeastlike endosymbionts and extracellular symbionts. Intracellular symbiosis Coleoptera - Anobiidae, Cerambycidae Homoptera - Delphacidae, Flatidae, Ricaniidae Extracelluar symbiosis Homoptera - Cerataphidini (Aphids) Hymenoptera - Encyrtidae

     The location of the symbiont within each host group varies.  In anobiid beetles, symbionts are located in the ceca of the larval midgut at the junction with the foregut.  During metamorphosis the ceca disintegrate, and the adult mycetome is formed within smaller ceca.  In adults, the symbionts are released into the intestinal lumen from the ceca and eventually reach the vaginal pockets.  Cells from the vaginal pockets are smeared on the eggs when they are oviposited.  Newly hatched larvae ingest the symbionts, which infect the ceca of the gut.  In addition to anobiid beetles, cerambycid beetles also harbor symbionts in ceca located around the midgut.  Transmission from adult to larvae is similar to that of the symbionts of anobiid beetles.  The larval ceca, however, disappear shortly before pupation, and in the female the symbionts multiply in the hindgut.
     Planthopper yeast-like symbionts are located within inner cells of the fat body. Soon after males emerge the number of symbiont cells decreases and they are lost gradually. In adult females some of the symbionts infect the ovarian epithelial plug (the part that connects the ovariole and pedicel) and enter the terminal oocytes situated at the most posterior part of the ovariole from the posterior pole.  Yeast-like symbionts are found in the egg by the time of shell formation.  Within the egg, they remain in a mass called the sy