Bark beetles (Coleoptera: Scolytinae) include many primary pests, which can cause significant economic losses to forests and forestry. The majority of these species are harmless to healthy living trees, infesting mainly dead or dying trees in their native environment (Paine et al. 1997, Martikainen et al. 1999, Knížek & Beaver 2004). An interesting characteristic of bark beetles is their widespread association with fungi; the most notable are the associations with ophiostomatoid fungi (Ascomycota) responsible for discoloration of wood and serious tree diseases (Wingfield et al. 1993, Kirisits 2004). Bark beetles are known to greatly facilitate the spread of these fungi. Both bark beetles and the fungi associated with them are easily transported through the movement of untreated wood products. Increased global trade in untreated timber and wood products raises the risk of accidentally introducing these forest pests and pathogens into a new environment (Tkacz 2002). Several examples of invasive bark beetle species and their associated fungi have shown that even species considered less harmful in their native environment can become potential threats to economic and ecologic well-being if accidentally introduced into a new environment (Ozolin & Kryokova 1980, Brasier 1983, Yin 2000, Li et al. 2001, Taylor et al. 2006, Lu et al. 2010). Considering the potential risks of introducing pests and pathogens in timber imported from Russia to Finland, a previous study identified a number of bark beetle species in the timber, including also potential pest species not native to Finland (Siitonen 1990). A changing environment can also increase the threats posed by these pests and pathogens (Williams & Liebhold 2002, Carroll et al. 2003, Berg et al. 2006). Although a number of studies have been devoted to resolving the nature of bark beetle-fungus interactions since they were first recognized in the 19th century (Schmidberger 1836, Hartig 1844, 1878), these interactions remain poorly understood. The studies regarding bark beetle-associated fungi are mainly focused on the economically most important bark beetle species. This might have biased the observations of true fungal biodiversity in the studied regions, and also our understanding of these symbioses (Six & Wingfield 2011). Not all bark beetle-fungus interactions should be viewed as one type of symbiosis having similar function. Apparently bark beetles and fungi form complex and dynamic associations, which are shaped during long periods of co-evolution and which are strongly influenced by the environment. The research concerning these fascinating symbioses is at the point where we are just learning to understand the diverse roles of fungi and their importance in the lives of bark beetles. 2.1 Taxonomy and morphology of ophiostomatoid fungi. Ophiostomatoid fungi represent an artificial group of fungi that consist of c.a. 200 species distributed in the Ascomycete genera Ceratocystis Ellis & Halst. (Microascales), Ceratocystiopsis H.P. Upadhyay & W.B. Kendr., Grosmannia Goid. and Ophiostoma Syd. & P. Syd. (Ophiostomatales). Adaptation to insect dispersal is typical for the majority of these fungi, and many of the species have a close association with their insect vectors (Wingfield et al. 1993). Ophiostomatoid fungi can be found on a wide variety of substrates in both the Northern and Southern Hemispheres. Due to the relatively simple morphology and overlapping features between different species, it has been difficult to identify these species, and their classification has been complicated and regularly revised. These confusing taxonomic debates have surrounded the ophiostomatoid fungi since the descriptions of the two major genera Ceratocystis and Ophiostoma. Phylogenetic studies based on DNA sequence data have clearly shown that despite the morphological and ecological similarities, these two keystone genera are phylogenetically unrelated and represent different orders of fungi (Hausner et al. 1992, 1993a,b, Spatafora & Blackwell 1994). Ceratocystis belongs to the Microascales together with related but economically unimportant genera, such as Gondwanamyces G.J. Marais & M.J. Wingf., Graphium Corda and Microascus Zukal. With the confusion between Ceratocystis and Ophiostoma resolved by modern taxonomic techniques, recent studies have focused on the Ophiostomatales. Recent DNA sequence analyses have defined three distinct phylogenetic lineages supported by morphological features in the Ophiostomatales: Ceratocystiopsis, Grosmannia and Ophiostoma (Zipfel et al. 2006). As the recent studies have demonstrated, DNA sequence-based identification has become essential for the reliable identification and recognition of cryptic taxa amongst these morphologically similar ophiostomatoid fungi (Gorton et al. 2004, Grobbelaar et al. 2009). Analyses of DNA sequence data have thus redefined the status of several genera and species and have led to the discovery of several previously unrecognized taxa. This is a trend that is likely to continue as more sequence data become available. Ophiostomatoid fungi have many morphological characters in common. These features are typically related to their adaptation for insect dispersal. The spore-bearing structures in both the teleomorph and anamorph states are in most cases long stalks, carrying the spores in slimy droplets. When possible, morphological identification has been based on the characteristics of both the anamorph and teleomorph structures. In many cases, the characterization is based on anamorph morphology only. Many species, particularly Leptographium Lagerb. & Melin spp., are not typically associated with a teleomorph, or the teleomorph is rarely observed. The typical teleomorphs of these fungi are characterized by globose ascomatal bases with elongated necks, evanescent asci and hyaline, one-celled ascospores having a wide variety of shapes (Hunt 1956, Upadhyay 1981, De Hoog & Scheffer 1984, Wingfield et al. 1993, Jacobs & Wingfield 2001, Zipfel et al. 2006). These fungi have a variety of different anamorphs, of which most also produce their conidia in slimy droplets. The sticky spore droplets can attach to the bodies of passing insects and thus facilitate the dispersal of the fungi. The morphological similarity of ophiostomatoid fungi is probably a result of convergent evolution, as adaptations to insect dispersal (Spatafora & Blackwell 1994).Species of Ceratocystis are characterized by Thielaviopsis Went anamorphs and endogenous conidium development (Halsted 1890, Minter et al. 1983). In contrast, conidium development of species in the Ophiostomatales is exogenic (Minter et al. 1982). Ceratocystiopsis is characterized by Hyalorhinocladiella H.P. Upadhyay & W.B. Kendr. and Sporothrix Hektoen & C.F. Perkins anamorphs, and small perithecia with long, falcate ascospores (Upadhyay & Kendrick 1975, Zipfel et al. 2006). At present, eleven species of Ceratocystiopsis are known. The species of Grosmannia are characterized by Leptographium anamorphs, and the presence of intron 4 and the absence of intron 5 in the β-tubulin gene (Goidánich 1936, Zipfel et al. 2006). At present, 28 teleomorph species are recognized in Grosmannia, with many more Leptographium spp. for which no teleomorphs are known. The remaining genus in the Ophiostomatales, Ophiostoma, is the largest, including more than 120 species and a variety of ascospore shapes and anamorphs in Sporothrix, Pesotum J.L. Crane & Schokn. and Hyalorhinocladiella, or combinations of these. The phylogenetic study of Zipfel et al. (2006) showed that the definition of Ophiostoma remains unsatisfactory. The study revealed that the genus is polyphyletic, forming lineages linked to morphological characters. Ophiostoma species with cylindrical or allantoid ascospores with pillow-shaped sheaths and a continuum of anamorphs, ranging from primarily Hyalorhinocladiella-type structures to more rare Pesotum-like synnematous structures, group with Ophiostoma ips (Rumbold) Nannf. and form the so-called Ophiostoma ips-complex (sensu stricto) (Zipfel et al. 2006). The species with relatively long allantoid ascospores and exceptionally long perithecial necks and Sporothrix anamorphs group within the Ophiostoma pluriannulatum-complex. The most challenging group to define is the Ophiostoma piceae-complex, which includes species with allantoid to cylindrical ascospores and a variety of anamorphs. This complex does not form a well-supported phylogenetic lineage. This is problematic, since the type species for Ophiostoma, Ophiostoma piliferum (Fr.) Syd. & P. Syd. falls in this group.There are no clear characters that can be used to define Ophiostoma as a distinct genus. Several phylogenetic studies have shown that the hardwood-infesting isolates group together (Harrington et al. 2001, De Beer et al. 2003b, Grobbelaar et al. 2009, Grobbelaar et al. 2010). The Sporothrix schenckii-Ophiostoma stenoceras -complex of species, characterized by reniform ascospores without a sheath, a Sporothrix anamorph (De Beer et al. 2003a), and the absence of intron 4 and presence of intron 5 in the β-tubulin gene (Zipfel et al. 2006), also represent a discrete group. The habitat of species belonging to this group is in contrast to other Ophiostomatalean species, which are associated with bark beetles or other tree-infesting insects. The majority of the species in S. schenckii-O.stenoceras-complex are found in soil. A recent study has shown that the species in this complex should be recognized as a distinct genus (de Beer et al. 2010). Also, the monophyly supported by the morphological and possibly ecological characters of the other emerging groups within Ophiostoma remain unresolved. This is likely to remain the case until sequences of more species and more genes clarify the genetic status of these complexes. Ophiostomatoid fungi also differ in the chemical composition of their cell walls (De Hoog & Scheffer 1984). The cell walls of Ophiostoma contain cellulose and rhamnose, which is unusual for the Ascomycetes. In contrast, the cell walls of Ceratocystis consist mainly of chitin. In addition, Ceratocystis and Ceratocystiopsis are very sensitive to the antibiotic cycloheximide, which inhibits the protein synthesis in most eukaryotic organisms (Harrington 1981, De Hoog & Scheffer 1984, Zipfel et al. 2006). Species of Ophiostoma are able to tolerate high concentrations of cycloheximide and this feature is commonly applied when these fungi are isolated from soil or insects. 2.2 Ecology of ophiostomatoid fungi. 2.2.1 Sapstain. Ophiostomatoid fungi are also known as "blue-stain fungi" or "sapstain fungi", referring to the bluish, grey, brown or black discoloration of sapwood caused by these fungi (Münch 1907, Seifert 1993). Other groups of fungi causing sapstain are black yeasts and dark molds (Seifert 1993). Sapstain-causing fungi are especially important in conifer trees in Northern Hemisphere (Seifert 1993, Butin 1996). The discoloration lowers the value of timber, but unlike the structural damage caused by soft-rot or decay fungi, the damage is mainly cosmetic. Staining is caused by fungal hyphae usually grown in the ray parenchyma cells and resin ducts (Münch 1907, Gibbs 1993, Seifert 1993). At later stages of infection, the tracheids are also colonized (Liese & Schmid 1961, Ballard et al. 1982). Discoloration is due to melanin, a pigment existing inside the fungal hyphae walls, and not to the staining of the wood tissues (Zink & Fengel 1989, 1990). 2.2.2 Plant pathogens. Several species of ophiostomatoid fungi are serious forest pathogens. The pathogenicity of these fungi has been demonstrated to vary greatly from weak pathogens to species capable of killing healthy trees (Horntvedt et al. 1983, Solheim 1988, Kile 1993). The best-known examples of the latter group are the Dutch elm disease pathogens, Ophiostoma ulmi (Buisman) Nannf. and Ophiostoma novo-ulmi Brasier, species responsible for the disastrous pandemics killing millions of elm (Ulmus L.) trees in both Europe and North America during the past century (Gibbs 1978, Brasier 1991, Hubbes 1999, Brasier & Kirk 2001). Other severe pathogens include the host-specific varieties of Leptographium wageneri (W.B. Kendr.) M.J. Wingf. causing black stain root disease in conifers in North America (Cobb 1988, Harrington 1993), Leptographium calophylli J.F. Webber, K. Jacobs & M.J. Wingf. causing Takamaka wilt disease (Ivory et al. 1996, Webber et al. 1999), and Leptographium procerum (W.B. Kendr.) M.J. Wingf. that has been associated with a disease known as white pine root decline, but most likely only contributes to the disease (Kendrick 1962, Alexander et al. 1988, Wingfield et al. 1988). Species of Ceratocystis are also causal agents of tree diseases, such as Ceratocystis fagacearum (Bretz) J. Hunt. causing oak wilt (Hepting 1955, Kile 1993) and members of the Ceratocystis fimbriata-complex causing canker stain and vascular wilt diseases in a wide range of host trees (Kile 1993). Several species of Ceratocystis are also economically important pathogens of food and crop plants. A recent review has summarized the current knowledge regarding diseases caused by Ceratocystis spp. (Roux & Wingfield 2009). 2.3 Interactions. Fungi are heterotrophs that depend on other organisms; therefore, different interactions are common and have developed early in the life history. To date, plant-fungi interactions are known to be older than interactions between fungi and insects (Taylor & Osborn 1996, Engel & Grimaldi 2004, Heckman et al. 2001). The terrestialization of the Earth by land plants might not have been possible without mutualistic plant-fungal interactions (Jeffrey 1962, Pirozynski & Malloch 1975). It has been hypothesized that the initial fungal associates of plants were saprobes with an invasive mycelium, having the ability to penetrate dying and dead cells (Taylor & Osborn 1996). As these plant-fungal interactions developed, fungi might have overcome the defensive mechanisms of plants, so that parasitic and eventually biotrophic interactions evolved. The earliest fungi were present in the Precambrian period (Heckman et al. 2001), and first examples of plant defense responses to fungal parasites come from the Devonian period (Taylor et al. 1992). While fungi and plants were forming symbiotic relationships at a very early stage in terrestrial evolution, insects had just originated in the Silurian period (Engel & Grimaldi 2004). None of the early insect fossils are known to have fungal associates (Taylor & Osborn 1996). Therefore, it can be assumed that fungi were first adapted to plants and that interactions between insects developed much later. Examples show that since these interactions started to develop, they have often led to complex and rather sophisticated associations (Hughes et al. 2010). The association between bark beetles and fungi was first recognized in the 19th Century (Schmidberger 1836, Hartig 1844, 1878). Due to the often destructive nature of these interactions, a number of studies have been devoted to resolving the nature of the associations. At present it is known that bark beetles, fungi and host trees form complex interactions, of which many are still only poorly understood. 2.3.1 Fungi-bark beetle interactions.Bark beetles are among the first insects that attack a dead or a weakened tree. They include species that reproduce in the inner bark (phloephagous species), and ambrosia beetles (xylomycetophagous species), which bore tunnels into the wood and cultivate and feed on symbiotic ambrosia fungi (Knížek & Beaver 2004). Bark beetle species are geographically widely distributed (Knížek & Beaver 2004), and occur in a wide range of host trees (Kirkendall 1983). In Nordic countries and Russian Karelia, entomological research has a long and intensive tradition, and the biology of forest pest fauna and their host range is well known (Lekander et al. 1977, Heliövaara et al. 1998, Mandelshtam & Popovichev 2000, Voolma et al. 2004). Probably due to the host choice behavior of the beetles, phloephagous species are normally specific to one tree genus, and only some species attack trees from closely related genera (Sauvard 2004, Bertheau et al. 2009). However, bark beetles are well suited for movement across national boundaries, and have adaptation capabilities that allow them to switch to novel host tree species if introduced to a new environment (Marchant & Borden 1976, Tribe 1992, Sauvard 2004, Yan et al. 2005). These potential new interactions are a matter of concern, since they can result in extensive insect outbreaks and damage in forest ecosystems. Most of the bark beetle species are harmless to healthy living trees, but some are regarded as important forest pests, causing significant economic losses (Knížek & Beaver 2004). Conifer bark beetle species are the most important forest pests in the temperate zones (Grégoire & Evans 2004). Bark beetle species that infest hardwood trees are considered less harmful, with the exception of the species vectoring the fungi responsible for the Dutch elm disease pandemics. In their native environment and in non-epidemic conditions, several bark beetle species are regarded as secondary, infesting dead or dying trees (e.g. Ips pini (Say), Scolytus ventralis LeConte (Paine et al. 1997, Martikainen et al. 1999, Knížek & Beaver 2004). They are organisms that have an important role in forest ecosystems accelerating the natural recycling of nutrients (Martikainen et al. 1999). Several bark beetles are keystone species driving forest succession, e.g. Ips typographus L. in Eurasia. A number of other organisms, such as arthropods and fungi, are associated with I. typographus (Weslien 1992, Viiri 1997). Bark beetle species can become economically important when they transfer pathogenic fungi to living trees, when their populations build to outbreak levels, or when they are introduced into new environments (Wingfield & Swart 1994, Knížek & Beaver 2004). A relatively small number are considered primary bark beetles (e.g. Dendroctonus mexicanus Hopkins and I. typographus) that attack living, healthy trees, seedlings or seeds of commercial crops (Coulson 1979, Wood 1982, Paine et al. 1997, Knížek & Beaver 2004). The majority of the bark beetle species have only minimal contact with living trees. Those species are saprophytes, which colonize only dead trees (Raffa et al. 1993, Paine et al. 1997). Ophiostomatoid fungi are common and relatively well-known associates of bark beetles (Münch 1907, Harrington & Cobb 1988, Wingfield et al. 1993, Paine et al. 1997, Jacobs & Wingfield 2001, Kirisits 2004). Ophiostomatoid fungi are commonly found in galleries constructed by bark beetles and their larvae in the phloem and wood of mainly coniferous trees (Kirisits 2004). Fungi sporulating in the galleries can be carried in mycangia, special organs of bark beetles (Francke-Grosmann 1967, Beaver 1989), attached to the surface of their exoskeletons (Beaver 1989), in the digestive tracts of the beetles (Furniss et al. 1990), or on mites phoretic on bark beetles (Moser et al. 2010). Usually bark beetles are associated with more than one fungus. Each bark beetle can transfer several fungal species, and thousands of conidia and ascospores, but great variation occurs between individuals (Solheim 1993a). The association of ophiostomatoid fungi with particular bark beetle species can be either specific or more casual. Bark beetle species with more casual associations can vector numerous fungi, but none of these fungal species is found consistently in high frequencies in bark beetle populations (Mathiesen-Käärik 1953, Solheim & Långström 1991, Gibbs & Inman 1991). For example, I. typographus is a vector of numerous ophiostomatoid fungi, of which many are reported only occasionally and in low numbers (Kirisits 2004). In specific associations between fungi and bark beetles, a large number of individual bark beetles regularly carry spores of certain ophiostomatoid fungi. Bark beetle species on hardwoods seem to more commonly have rather fixed associations, e.g. Scolytus Geofroy spp. infesting elms are regularly associated with the Dutch elm disease fungi. Studies of beetle-associated flora are generally focused on reporting the fungal associates of different bark beetle species. Lieutier et al. (2009) suggested that the host tree has a more important role than the beetle in the speciation of ophiostomatoid fungi. In the evolutionary sense, plant-fungi interactions are known to be older than interactions between fungi and insects (Taylor & Osborn 1996, Engel & Grimaldi 2004, Heckman et al. 2001). Studies regarding the origin of associations between ophiostomatoid fungi, the host tree and the vector insect are lacking. In the light of knowledge from plant-fungi interactions in general, it is possible to conclude that the adaptation of ophiostomatoid fungi to trees is also older than their adaptations to bark beetles (Harrington & Wingfield 1998, Lieutier et al. 2009).Interactions between bark beetles and their fungal associates are diverse, ranging from antagonism and commensalism to mutualism (Klepzig & Six 2004). In many cases, the symbiosis is thought to be mutualistically benefitting for both the beetles and the fungi (Francke-Grosmann 1967, Beaver 1989, Berryman 1989, Ayres 2000). The dispersal of the ophiostomatoid fungi almost completely depends on the insect vectors, and therefore the fungi benefits from the association with the beetle vectors by transport to new host trees (Dowding 1969, Paine et al. 1997, Klepzig & Six 2004). Ophiostomatoid fungi have evolved adaptations to facilitate this transfer between trees. The fruiting structures of ophiostomatoid fungi are usually long stalks bearing spores in slimy droplets and concave shapes to allow multiple contact points, which can easily attach to the surface of the insect vector (Malloch & Blackwell 1993). Sticky ascospores ensure that they adhere tightly to the body of the insect and disperse in the resin of the new host, not the water (Whitney & Blauel 1972). Besides rapid transport to a suitable habitat, insect dispersal provides protection from desiccation and UV light (Klepzig & Six 2004). For some mutualistic fungi, sexual recombination has become apparently disadvantageous, and they lack or rarely possess sexual reproduction (Wulff 1985). These morphological features are considered as adaptations to insect dispersal and to the bark beetle habitat (Francke-Grosmann 1967, Whitney 1982, Beaver 1989, Malloch and Blackwell 1993). The evolution of mycangia, the special spore-carrying structures of bark beetles, indicates that some beetles also benefit from the association with fungi (Paine et al. 1997, Harrington 2005). In their nutrition-poor substrates of wood tissues, some bark beetles are dependent upon their fungal associates as a source of nutrients, or benefit from feeding on the fungi (Beaver 1989, Fox et al. 1992, Kopper et al. 2004). Female ambrosia beetles carry the primary fungus in the mycangium, often together with an assemblage of other fungi, yeasts and bacteria (Batra 1966, Haanstad & Norris 1985). In the new host tree, bark beetles plant and tend the primary fungus in their galleries (Norris 1979). The ways bark beetle species benefit from the association with fungi include feeding on the fungi, modifying the substrate to be more suitable for the larval diet providing compounds such as nitrogen, sterols and proteins, and by limiting the growth of harmful fungal species (Beaver 1989, Paine et al. 1997, Ayres et al. 2000, Klepzig & Six 2004). Besides the apparently positive benefits to bark beetles, some ophiostomatoid fungi are antagonists of bark beetles by negatively affecting the development of these insects (Kirisits 2004). Studies regarding antagonistic effects conducted in North America found that Ophiostoma minus (Hedgc.) Syd. & P. Syd. can negatively affect the development of Dendroctonus frontalis Zimmermann in North America (Barras 1970, Klepzig et al. 2001), and O. ips affects Ips spp. in a similar way (Yearian et al. 1972). The same fungus can have both positive and negative affects to its symbiotic bark beetle vector. It has been suggested that the reason for this apparently mutualistic interaction might be the importance of O. minus in the tree-killing process of D. frontalis (Nelson 1934, Bramble & Holst 1940, Mathre 1964). After O. minus has successfully been introduced into the host tree, it competes with the bark beetle for the uncolonized tissues (Klepzig & Wilkens 1997).The possible benefits of fungal associates to bark beetles in the process of successful colonization of living trees have been the subject of continuing debate. Several bark beetle-associated fungi have been considered to facilitate the bark beetle colonization by helping to overcome host resistance and killing the tree (Nebeker et al. 1993, Paine et al. 1997). This classic paradigm (CP) suggests that many bark beetle-fungus associations are mutualistic, based on the phytopathogenicity of the fungal associates (Six & Wingfield 2011). The results of several studies focused on these host tree-bark beetle-fungi interactions have been controversial and without conclusive evidence to support the CP. 2.3.2 Fungi-host tree-bark beetle interactions.Our current understanding of the interactions between bark beetles, fungi and host trees is insufficient and thus beset with controversy. Here I will discuss only a few aspects of the presented arguments. The varying aspects have been discussed in more detail in several articles (e.g. Whitney 1982, Harding 1989, Raffa & Klepzig 1992, Harrington 1993, Paine et al. 1997, Lieutier 2002, 2004, Lieutier et al. 2009, Six and Wingfield 2011), and the debates will certainly continue in future. In Fennoscandia, the dispersal and the host finding phase of the bark beetle life cycle is in May-June (Saalas 1949, Heliövaara et al. 1998). Bark beetles overwinter in the forest litter or under the bark of trees and begin their dispersal flight to seek suitable host trees in which to reproduce (Byers 1996). Bark beetles locate the suitable host tree by random landing and testing the tree and its resistance capability (Moeck et al. 1981, Wood 1982). Bark beetles have a pheromone-phased communication system that helps them to select and colonize suitable host trees (Moeck et al. 1981, Bakke 1983). After the selection of the host tree, bark beetles release pheromones that attract mates and additional colonists, leading to a rapid aggregation of a large number of beetles in the potential host tree (Raffa & Berryman 1983). Mutualistic relationships between phytopathogenic fungi have been proposed to be essential for bark beetles to successfully colonize living trees (Francke-Grosmann 1967, Graham 1967, Raffa & Berryman 1983). The tree killing hypothesis suggests that virulent fungi are responsible for tree death by blocking water conduction in the colonized tree (Långström et al. 1993, Paine et al. 1997). According to another hypothesis, fungi cause tree death indirectly by stimulating induced defense mechanisms of the host tree (Lieutier et al. 2009). Since the early propositions, the assumption was for many years that fungi are responsible for killing the trees attacked by bark beetles before the bark beetles can successfully continue the colonization (Berryman 1982, Coulson 1979, Wood 1982). The importance of ophiostomatoid fungi in host tree infestation by bark beetles has been tested using studies on fungal-free progenies of bark beetles. It has been demonstrated that the presence of ophiostomatoid fungi is not a prerequisite for successful the reproduction of some bark beetle species (Grosmann 1931, Harding 1989, Colineau & Lieutier 1994). Additionally, tree-killing bark beetles are able to kill trees without virulent fungal associates (Hetrick 1949, Bridges et al. 1985).The role of fungi associated with bark beetles has been demonstrated in a number of studies attempting to mimic bark beetle attacks by artificially inoculating living host trees with symbiotic fungi (e.g. Christiansen 1985, Solheim et al. 1993, Yamaoka et al. 1995, Krokene & Solheim 1998, Kirisits 1998). The lesion length caused by the fungal infestation has been used as a measure to study the virulence of a fungus (Matsuya et al. 2003, Rice et al. 2007). Under the tree killing hypothesis, the most virulent fungal associates are believed to be the most effective in killing the tree, and therefore the most useful for bark beetles (Yamaoka et al. 1990, Solheim & Safranyik 1997). The defense exhaustion hypothesis suggests that the most virulent fungal associates are the most effective in exhausting tree defense mechanisms. Studying these hypotheses included in the CP has several difficulties, and the results from the studies have been controversial. However, the CP has strongly influenced the research on bark beetle-fungus symbiosis during the last decades. Recently, the CP has been proposed to be fundamentally flawed (Six & Wingfield 2011). Six & Wingfield (2011) suggest that fungal phytopathogenicity has a more important role for the fungi, rather than supporting the bark beetles in tree killing. Fungal pathogenicity may be a factor helping the fungi to survive in a living tree (Six & Wingfield 2011). Pioneer fungal species need to be able to colonize tissues that are still living, or be able to tolerate the defensive reactions of trees formed in response to the beetle attack. Highly virulent fungi might need to be able to survive in a living tree, because they live in association with bark beetles completing their entire life cycle in living trees (Six & Wingfield 2011). Fungi that do not have virulent properties might be those infesting trees later. For example, species such as Ceratocystis polonica (Siemaszko) C. Moreau, considered highly pathogenic in artificial inoculation studies, are the first species that invade sapwood (Solheim 1993a). Typical of these species is the fact that they have rapid growth rates and tolerance to low oxygen levels.2.3.3 Fungi-fungi interactions. One relatively well-known example of fungi-fungi interactions is the interaction between mycangial fungi and other fungi. Fungi carried in the mycangia of ambrosia bark beetles compete with other fungi carried by the beetles, and can affect the fitness of bark beetles by limiting growth of co-occurring fungi (Norris 1979, Mueller et al. 2005). Ambrosia beetles carry one primary fungus intended for cultivation, and the other fungi are possible weeds that soon contaminate and overrun the cultivated fungal gardens if they remain untended. Mycangial fungi are considered less-virulent species (Paine et al. 1997).Trees attacked by bark beetles are subjected to colonization by several fungal species competing for the same resources. Ophiostomatoid fungi are known to be more tolerant to terpenes in conifer resin than other co-occurring early colonizing fungi, and thus some species may actually benefit from these defense reactions in the competition with other fungi (Cobb et al. 1968, DeGroot 1972, Harrington 1993, Klepzig & Six 2004, Lieutier et al. 2009). Competition between pioneer fungi, including interspecific competition between ophiostomatoid species, might play an important role in the successful colonization and pathogenic properties of fungal species (Owen et al. 1987, Parmeter et al. 1989, Harrington 1993). Bark beetles typically have multiple fungal associates. If competition between fungal symbionts is the only mechanism shaping the bark beetle-fungus interactions, there would be a strong evolutionary selection pressure driving the selection of the most competitive fungal associate (Six & Wingfield 2011). One hypothesis for the occurrence of multiple fungal associates at the same time is that although the fungi seem to occupy the same niche, separation into niches actually exists. This separation into niches reduces competition and thus allows the coexistence between several fungi. The niche separation might be a result of different temperature tolerance; resource use, such as the use of carbon and nitrogen sources; and a different degree of virulence between fungi (Six & Paine 1997, Solheim & Krokene 1998, Bleiker & Six 2007, Six & Wingfield 2011).2.4 Impact of globalization and environmental change.The majority of the bark beetle species are considered rather harmless species, colonizing mainly weakened or dead trees. However, these species pose potential risks in changing or new environments. Therefore, they should not be ignored when evaluating risks and threats to economic and ecologic well-being or when determining quarantine measures for pests and pathogens. Forest pest insects and their associated micro-organisms are capable of movement through national boundaries. International trade and travel between and within continents has increased the rates of these forest pest introductions to new environments. For example, a recent study has listed 109 exotic phytophagous insect species originated from North America and Asia that successfully invaded and established themselves on Europe's woody plants (Vanhanen 2008). The risk of successful establishment in a new environment is most severe when the main host species for the introduced pest species occurs naturally or is also introduced and widely cultivated. Changes in the climate might also induce invasions of both native and exotic insect pests from southern locations to northern locations, and increase the frequency and intensity of forest insect outbreaks (Ayres & Lombardero 2000). For example, a temperature increase can significantly affect the reproduction and population dynamics of I. typographus in Northern Europe (Jönsson et al. 2007). A classic example of the impact of invasive species is found in the Dutch elm disease fungi. It has been hypothesized that the Dutch elm disease fungi were originally native to the Himalayas (Brasier 1993), from where the pathogen was accidentally introduced into America and Europe. Elm species in America and Europe did not have resistance to the pathogen (Ozolin & Kryokova 1980, Heybroek 1981), which resulted in two destructive pandemics wiping out millions of the elm trees.There are also several current examples of the major devastation that bark beetles and their fungi caused as a result of environmental changes or where they have been introduced into foreign environments. One example is the mountain pine beetle (MPB) epidemics. The mountain pine beetle (Dendroctonus ponderosa Hopk.) is native to pine forests in western parts of North America. It primarily infests lodgepole pines (Pinus contorta Dougl. Ex. Loud.), but can infest most pine species occurring in the region. The Lodgepole pine is widely distributed in Canada, and therefore the occurrence of the beetle species in the western Canada is not restricted by the availability of a suitable host tree. Climate is the major factor limiting the MPB to expand to northern and eastern parts of Canada (Safranyik 1978). Normally the MPB infests weakened and dying trees. However, periodical large-scale outbreaks in healthy trees are also normal behavior of the MPB (Safranyik & Carrol 2006). Current epidemics in British Columbia, Canada are more severe and larger in area than any of the previous outbreaks recorded (Taylor et al. 2006). The epidemic is occurring in areas previously considered climatically unsuitable for the MPB (Safranyik 1975). This shift to formerly climatically unsuitable areas during the last several decades has been explained by climate change and global warming. The sufficient changes in the climatic conditions, such as increased temperatures and reduced summer precipitation have allowed mountain pine beetles to establish and form self-sustaining populations in the new areas (Williams & Liebhold 2002, Carrol et al. 2003). Another example of a bark beetle outbreak - induced by climate change which has led to significant damage in North America, Alaska, is the spruce beetle (Dendroctonus rufipennis Kirby) (Berg et al. 2006). As a result of increased temperatures, the reproduction time of the spruce beetle has halved and led to extensive and unprecedented damage to spruce forests. An example of a beetle and its associated fungi recently introduced into a new environment is the red turpentine beetle (Dendroctonus valens LeConte). The red turpentine beetle is normally a non-aggressive species infesting dead or dying conifers, mainly Pinus ponderosa Dougl. ex Laws. in North America (Smith 1961). The red turpentine beetle was introduced from the pine forests of North and Central America to China around 1980 (Pajares & Lanier 1990). In China, it spread rapidly since the first outbreak in 1999, causing significant damage in over half a million hectares of pine stands (Yin 2000, Li et al. 2001, Miao et al. 2001). In China, the main host tree species for D. valens is Pinus tabuliformis Carr. (Li et al. 2001). The red turpentine beetle vectors an ophiostomatoid fungus, Leptographium procerum (W.B. Kendr.) M.J. Wingf., which is non-pathogenic in the USA, but has become a serious pathogen of pine in China (Lu et al. 2010). The invasive strains of the fungi tolerate higher concentrations of monoterpenes and are thus better adapted to the host's defense response. There is also evidence that the fungus may increase beetle fitness by increasing the weight of the larvae that feed on the fungus.Numerous contemporary examples illustrate that bark beetles previously considered minor pests can become substantial threats in changing or new environments. Thus all bark beetle species and the fungi they carry should be considered as potentially threatening. This is at least within the context that they may not necessarily behave similarly in their native and introduced ranges. 2.5 Occurrence of Ophiostoma spp. and Grosmannia spp. in Fennoscandia.Previous studies have recorded 16 species of Ophiostoma and 13 species of Grosmannia and its anamorph Leptographium spp. occurring in association with pine-, spruce- and birch-infesting bark beetles in Fennoscandia (Table 1). The investigations thus far have included 15 bark beetle species, of which 14 infest conifers and one infests hardwood species. The most extensively studied bark beetle species is I. typographus. The investigations conducted in larger part of Europe have recently been reviewed by Kirisits (2004).The diversity of ophiostomatoid fungi that bark beetles vector in Fennoscandia shows differences compared to southern parts of Europe. The species diversity is seemingly lower in northern parts of Europe. Several species that have been reported in other parts of Europe have never been detected in studies conducted in Fennoscandia. Several species have also been regarded as more common associates in northern parts of Europe, including species such as C. polonica, Grosmannia penicillata (Grosman) Goid., Ophiostoma piceae (Münch) Syd. & P. Syd., Grosmannia piceiperda (Rumbold) C. Moreau, O. minus, Ophiostoma ainoae H. Solheim and Ophiostoma bicolor R.W. Davidson & D.E. Wells. However, the studies on ophiostomatoid fungi especially in Finland and neighboring Russia are limited. Reports of Ophiostoma and Grosmannia species from Russia are more numerous, but to our knowledge, none of the studies is conducted in Fennoscandian parts of Russia. The majority of the studies in Russia have focused on middle Siberia and southeastern parts of the vast country. Bark beetles and host trees that are common in the boreal forest of Siberia are not widely distributed in the European parts of Russia. The distribution of bark beetles that are considered quarantine pests in Europe, Ips cembrae Heer and Ips subelongatus Motschulsky, follows the distribution of larch (Larix Mill. sp.) (Stark 1952). However, several ophiostomatoid species reported from conifer-associated bark beetles from Siberia are also typical to Fennoscandia. These include O. ainoae, O. bicolor, O. minus and O. piliferum (Pashenova et al. 1995, 2004). On the contrary, although elms (Ulmus spp.) occur in southern Finland and parts of Russian Karelia, none of the elm-infesting Scolytus spp. have been found in this region (Jakovlev & Siitonen 2005). There are also no reports of the occurrence of species responsible for the Dutch elm disease from Fennoscandia.Studies regarding bark beetle associated fungi in Fennoscandia are rather limited and their main focus on the fungal associates of aggressive bark beetles might have biased the true fungal biodiversity in the region. Based on previous studies, boreal forests in Fennoscandia and Russia seem to have rather similar bark beetle associated fungal flora, which have differences compared to forests in southern parts of Europe. This difference can be explained by climate and the general migration pattern of taxa in Northern Europe, strongly affected by periods of glaciations (Hewitt 1996). For example, recent molecular analyses and fossil records have revealed that the Norway spruce populations in Northern and Central Europe form two distinct lineages, which have been isolated from each other for a long time (Tollefsrud et al. 2008). The populations in Northern Europe have originated from Russia, and spread from there to Scandinavia. The distribution of bark beetles is known to follow the distribution of their host trees. Therefore, it can be assumed that also the fungi associated with the bark beetles have followed the same distribution routes, forming rather similar flora in Northern Europe, which substantially differ from the flora e.g. in Central Europe. 2.6 The fungal species concept. The concept of species is ambiguous in mycology. Species is commonly used as the basic rank in taxonomy, but what is considered to be a species can vary widely (Guarro et al. 1999). Asexual reproduction and hyphal anastomosis are widely spread characters in fungi, and therefore an individual is not always easy to distinguish from a population in mycology (Carlile et al. 2001). Attempts to create a universal definition of species have failed, and are most likely bound to remain unresolved (Hey 2001). Thus several different approaches for delineating the species have been used. The most widely accepted approaches are the morphological species concept, the biological species concept, and the phylogenetic species concepts (e.g. Guarro et al. 1999, Taylor et al. 2000). The classic and most widely used concept by mycologists has been the morphological species concept. The ca. 100 000 identified fungi are mainly described by morphological characters (Kirk et al. 2008). The weakness of the applications of this concept is that diagnosed species often comprise more than one species, and it cannot be counted on to diagnose evolutionary meaningful species of fungi (Taylor et al. 2000). The morphological species concept cannot be used as the only approach to ophiostomatoid fungi because of their relatively simple morphology and overlapping features between different species. Since the monograph by Upadhyay (1981), the taxonomic understanding of ophiostomatoid fungi has improved, and it is now clear that several morphology-based species descriptions are too broad. In many cases, the biological species concept is less ambiguous in mycology and has held a prominent place in species recognition (Taylor et al. 2000). Ernst Mayr (1970) defined that "species are groups of interbreeding natural populations that are reproductively isolated from other such groups." In ophiostomatoid fungi, the biological species concept was first applied to show the species delimitations in the Ophiostoma ulmi-complex (Brasier 1986), and the morphologically similar O. piceae-complex (Brasier & Kirk 1993, Halmschlager et al. 1994, Pipe et al. 1995). A serious problem with fungi, including ophiostomatoid fungi, is that mating tests are not possible to apply to fungi if they lack meiospores, are homothallic, cannot reproduce in cultivation or cannot be cultivated (Reynolds 1993). According to Taylor et al. (2000), an even deeper problem with the biological species concept is that fungi can be genetically isolated in the nature, but retain the ancestral character of interbreeding. The phylogenetic species concept has been used increasingly. Phylogenetic approaches and analyses of DNA sequence data have helped in resolving confusing taxonomic debates and they have greatly increased the taxonomic understanding of fungi (e.g. Zipfel et al. 2006, Grobbelaar et al. 2010). However, the definition of the phylogenetic species concept is not without complications. Ranking individuals in order to determine whether they can be considered different species by using phylogenetic analysis of a single gene without including additional information, such as mating tests, is uncertain due to the possibility of polymorphism (Taylor et al. 2000). The subjectivity can be avoided by using comparisons of more than one gene genealogy a phylogenetic approach called genealogical concordance phylogenetic species recognition (GCPSR). Recent studies of ophiostomatoid fungi have widely adopted multi-gene phylogenies in combination with an evaluation of morphological and biological characteristics (Zipfel et al. 2006, Grobbelaar et al. 2010). 2.7 Tools for molecular identification. For phylogenetic species recognition, the genes encoding nuclear and mitochondrial ribosomal RNA genes and associated spacer regions are widely used. The universal and conserved nature of these genes makes them useful in studying fungi, as well as plants and animals (White et al. 1990). The nuclear ribosomal genes are relatively easy to study, because they are arranged in long tandem repeats, which means the gene is already amplified in the genome and only a moderate amount of initial template DNA is needed. The nuclear large subunit (LSU; 28S or 25S) and the mitochondrial small and large subunit genes (SSU, LSU) are used at intermediate taxonomic levels, e.g. to show the position within the genus and the order (Geiser 2004). Analyses of SSU and LSU have been used to identify major monophyletic groups and to suggest their branching orders (Sogin et al. 1996). Nuclear small ribosomal RNA subunit genes, including the 18S gene and internal transcribed spacer regions (ITS) have widely been used for species level studies in many fungi. The three subloci of the ITS regions have different rates of evolution: a highly variable ITS1, a very conserved 5.8S gene and a variable to a semi-conserved ITS2 (Hillis & Dixon 1991, Hershkovitz & Lewis 1996). The two spacer regions (ITS1 and ITS2) are transcribed but do not encode a gene product, and thus evolve faster than the ribosomal subunit genes (Geiser 2004). The use of the ITS sequence for species level studies is sometimes problematic, because in some groups of fungi, ITS sequences have been observed to be either too variable to determine a major group (Der Bakker et al. 2004), or too conserved to distinguish between species (Du et al. 2005). Therefore, the ITS region is commonly used as the first step in molecular identification of fungi, and in several cases, another gene or genes are needed for a precise identification. Depending on the fungi, the uses of additional genes need to be determined. In studies of ophiostomatoid fungi, the ITS region was successfully used to resolve the confusing taxonomy between Ceratocystis and Ophiostoma (Hausner et al. 1993a, b, Spatafora & Blackwell 1994). Between more closely related species, the ITS region is sometimes too conserved and fails to separate very closely related phylogenetic species, such as O. piceae and O. canum (Harrington et al. 2001). In recent studies, the use of ITS sequence data together with protein-coding genetic data, such as β-tubulin and translation elongation factor 1-α (EF1-α), have become the norm (Lim et al. 2004, de Meyer et al. 2008, Roets et al. 2008, Grobbelaar et al. 2010). These intron-rich, highly conserved genes provide more resolution at the species level identification than the ITS region (Geiser 2004). The introns of these protein-coding genes evolve at a higher rate than the introns of the ITS region. 3 AIMS OF THE PRESENT STUDY. Despite the economic and ecological importance of forests in Finland, there is very little information on the occurrence of ophiostomatoid fungi in the commercially important tree species. Basically, previous studies have listed the fungal associates of the spruce bark beetle, I. typographus (Savonmäki 1990, Viiri 1997, Ahtiainen 2008). Apart from these studies, almost nothing is known regarding fungal associates of other less-aggressive bark beetle species native to the region. For approaches to understand bark beetle-fungus interactions better there's a need to study interactions that include also other than economically important bark beetle species (Six & Wingfield 2011). Most bark beetle species are known to carry spores of the ophiostomatoid fungi, and are easily moved through the movement of untreated wood products. Depending on the habits of their bark beetle vectors, these fungi can cause damage either on trees, logs or lumber. The fungi can weaken or kill trees and/or decrease value of the wood due to sapstain. Also, the presence of these insects and fungi in the imported lumber raises concerns in countries importing wood products, especially if they do not naturally occur in the importing country. Finnish forests cannot supply the demand of the industry and the country relies heavily on Russia as the source of raw timber. The risks of introducing pests and pathogens are difficult to assess because there are only limited studies concerning the possible pest insects and pathogenic fungi on timber imported from Russia to Finland. For example, a large number of bark beetle species, including potential pest species not native to Finland, were identified by Siitonen (1990). However, studies on the fungal associates of these beetles in Finland and Russia are limited and the species identification was made at a time when DNA sequence comparisons were not commonly applied. Furthermore, we are not aware of any study where the bark beetles and their fungal flora in the two countries have been compared. The general objective of this study was to provide more information on associations between fungi and bark beetles, including both aggressive and non-tree-killing bark beetles. The first aim was to isolate ophiostomatoid fungi associated with the most common bark beetle species infesting the dominant and commercially most important tree species in the boreal forests, Norway spruce, Scots pine and birches (Betula pendula Roth, Betula pubescens Ehrh.). The second objective of this study was to identify all ophiostomatoid fungi collected during this survey using morphological characteristics and DNA sequence comparisons for the rDNA gene regions and part of the β-tubulin and EF-1α genes. During the study, we found that the number of ophiostomatoid species is far more than previously thought. Also, a need for redefining some taxonomic groups emerged in this study. Therefore, some of the groups originally intended to be included in this study were left for future investigations. These taxonomic issues are discussed in the summary. The focus of this study is to report species of Ophiostoma and Grosmannia, which are the economically most important genera reported in several previous studies. Several new species of Ophiostoma and Grosmannia were found among the obtained isolates, which raised the third objective of describing these newly discovered species. The final objective was to compare the species diversity in Finland and Russia, and assess possible risks involved in the timber import. The fungi included species from different groups within ophiostomatoid fungi, which causes problems in DNA sequence alignment. Therefore, we decided to publish the results in smaller and more meaningful parts. Studies I-II present the bark beetle and fungal associates of hardwood trees (birches). In study I, we reported the fungal associates of birch bark beetles and described one new, a seemingly strict fungal associate. In study II, the ecology and distribution of this newly described species was further investigated. The study also revealed a new species, which was formally described. Studies III-IV present the fungal associates of conifers (pine and spruce). Study III presents all species of Ophiostoma associated with pine and spruce-infesting bark beetles, and includes a description of five new Ophiostoma spp. In study IV, we report all species of Grosmannia associated with pine and spruce-infesting bark beetles, and describe one new species. 4 MATERIALS AND METHODS. 4.1 Bark beetle occurrence and identification. The bark beetle collections for this study were mainly obtained from eastern parts of Finland and Russian Karelia in July 2005. Additional collections were obtained from Russian Karelia in June 2004 and July 2007, and southern Norway in July 2007. The landscape of the main study region is covered by mainly coniferous boreal (taiga) forests fragmented by open mires and lakes, cultivated land and sparse settlements. The climatic conditions in the region are characterized by warm summers and cold winters. Biogeographically, all the studied regions in this study (Finland, Russian Karelia and Norway) belong to the same area, Fennoscandia. During the last glaciations, Fennoscandia was covered by an ice sheet that started to retreat about 10,000 years ago. The tree species and their associated pests spread along the same postglacial routes from east-southeast. Therefore, the tree-species composition in the area is rather uniform; the dominant tree species in the study area are Scots pine, Norway spruce and birches (B. pendula and B. pubescens).The largest collections of bark beetles were obtained in eastern Finland and Russian Karelia in July 2005 (Table 2). Despite the geographical closeness, the forestry practices differ distinctively on the different sides of the border, between Finland and Russia. Forests in Finland have been intensively harvested over the last decades, while on the Russian side of the border forests have not been subject to similar intensive forestry management, and relatively high proportions of patches close to their natural state can still be found. Also, higher amounts of dead wood can be found from the forests of Russian Karelia. The proportion of dead wood can have an effect on the bark beetle and fungal populations (Martikainen et al. 1996). These features make the region ideal for studying species diversity in both areas of intensive forestry practices and areas where human influence has been slight. During the flight period of bark beetles in the region (Heliövaara et al. 1998), freshly cut trapping logs were laid on the forest floor to allow for their natural colonization. Different bark beetles and their galleries were collected from birch, spruce and pine trapping logs and/or naturally infested trees at four different sites in Finland and seven in Russia. Approximately 600 hectares of spruce-dominated forest was felled by a storm in Lake Vodla national park in Russian Karelia during summer 2000 (Roininen et al. 2005). The majority of these storm-felled trees remained in the forest floor. These storm-felled trees were soon mass-attacked by I. typographus, and in autumn 2003, large amounts of healthy standing trees attacked and killed by I. typographus were also recorded. Large areas of killed standing trees were also observed outside the storm-felled region. The I. typographus samples collected by H. Roininen from this extensive spruce bark beetle damage area in the Ohtama and Pilmazero regions of Russia are included in this study (Table 2). Also, collections of Tomicus piniperda L. from Volosovo region in Russian Karelia in July 2007 obtained by E. Sidorov are included (Table 2). Bark beetles were collected from felled pines in a pine stand that was 80-100 years old. An additional collection of Scolytus ratzeburgi Jans. infesting birch in southern Norway was obtained. The sampling was conducted in July 2007 in Akerhus and Østfold counties, Norway. 4.2 Fungal isolation and identification. Ophiostomatoid fungi were isolated from bark beetles, as well as from their galleries on Betula spp., Picea abies and Pinus sylvestris. Also, these fungi can be isolated from other insect vectors, lesions caused by the fungi and from soil around roots of diseased trees (Jacobs & Wingfield 2001). To prevent the colonization of secondary fungi, samples should be processed soon after they are collected. The bark beetle galleries were placed in humid chambers and incubated in room temperature for several weeks to allow fungi to sporulate. The humid chambers consisted of a Petri dish containing moisture paper. During this incubation period, the mycelium and/or fungal spore masses that formed in the galleries were detected under a dissecting microscope (32 ´ magnification). A sterile needle or fine sterile forceps were used to isolate fungi from aerial mycelium, from masses of spores on perithecia as well as from mononematous and synnematous conidiophores. The samples were transferred to 2 % malt extract agar (MEA) containing 0.05 % cycloheximide or streptomycin to prevent bacterial growth. Male and female beetles were squashed and streaked on the surface of the same media. Malt extract agar is selective for Ophiostoma species, and it often results in a good sporulation. Species of Ceratocystis are sensitive to cycloheximide (Seifert et al. 1993). Therefore, MEA containing streptomycin was also used in this study, but to a lesser extent. Once the resulting fungal isolates were purified, they were grouped based on the morphology. The isolates having similar aerial mycelium, mononematous of synnematous conidiophores, growth rate, growth pattern, colony margin and color were grouped together, representing potentially the same fungal species. All isolates were transferred also to
oatmeal agar (OA) and to MEA, to which sterilized birch, pine or spruce (depending on the origin of the strain) twigs were placed. Both media are suitable for some fungi to sporulate well (Seifert et al. 1993).4.2.1 Morphological characteristics. The cultural characteristics of species described in the study are based on the colony description of the representative isolates grown in 20 oC. The colony colors are defined with according to Rayner's (1970) color chart. The microscopic characteristics were examined using a phase contrast microscope (Nikon Corporation, Tokyo, Japan). For the species descriptions, anamorph and teleomorph (where present) fruiting structures were mounted in 85 % lactic acid on glass slides and examined with a 10Ã-, 25Ã-, 40Ã- objective or a 100Ã- oil-immersion objective. Measurements were made of 50 of each of the relevant morphological structures so that the ranges and averages could be computed. The 0.5 mm scale was used in studies I-IV (the theoretical resolution for a light microscope is 0.2 mm). The photographic images were captured with A Nikon DS-F11 camera system (Nikon Corporation, Tokyo, Japan). 4.2.2 DNA sequence data. For molecular identification, one isolate from each morphological group was chosen for DNA extraction and sequencing. The widely sequenced DNA region in fungi, the internal transcribed spacer (ITS) region, was chosen as a keystone for molecular identification. NCBI BLAST searches were conducted for the preliminary identifications. The GenBank database provides an increasing number of fungal ribosomal sequences, particularly for the ITS region (Geiser 2004). A fungus can be identified reasonably well at least to genus level by submitting its ITS sequence and performing a BLAST search in the GenBank. Based on the BLAST results and the preliminary phylogenetic analyses using ITS sequences, the fungi were further grouped and the need for further sequencing was determined. In several cases, the ITS region did not resolve phylogenetic species very well. Therefore, protein coding genes, β-tubulin (partial gene) and in some cases EF1-α (partial gene), were sequenced to provide more resolution and to confirm the results obtained from the analyses of the ITS region. Based on the BLAST results and the preliminary phylogenetic analysis, some sequences were highly novel, representing previously undescribed taxa. Some of those possible new species were also subjected to sequencing the LSU gene to show their placement at higher taxonomic levels. Also, more isolates representing different species were chosen for ITS region sequencing. When possible, isolates were selected for DNA sequencing to represent as wide ecological and morphological variation as possible; different locations, host trees, bark beetle vectors and morphological groups. 4.2.3 Phylogenetic analyses. The BLAST searches were conducted for the preliminary identifications, after which datasets were compiled that included reference sequences from GenBank. All datasets were compiled and the preliminary phylogenetic analyses were done using Molecular Evolutionary Genetic Analyses (MEGA) v3.1 (Kumar et al. 2004). Phylogenetic trees used in the articles were also edited using MEGA. Before the phylogenetic analyses, the datasets must be aligned. The compared sequences usually have different lengths, which means the locations of insertions and deletions must be inferred by introducing gaps in the DNA sequence alignment (Nei & Kumar 2000, Salemi & Vandamme 2003). In the multiple sequence alignment, the idea is to identify homologous regions within several related sequences. Divergent sections in sequences are sometimes problematic in multiple sequence alignments. The error rate in the alignment increases as divergence increases, and can cause the related part of the sequences to show lower similarity than they actually have. This is a problem also within ophiostomatoid fungi, which comprises fungi distributed to different genera. Based on preliminary phylogenetic analyses, isolates could be designated into different genera and complexes of species amongst them. Therefore, separate analyses of sequences for isolates representing taxa in different genus and also different complexes of species were necessary because of their differences in the presence and absence of introns. All datasets in studies I-IV were aligned using the online version of MAFFT v6 (Kumar et al. 2002). MAFFT is a fast and accurate multiple sequence alignment program, which has achieved the best results in alignment accuracy in a comparison of several multiple alignment programs (Nuin et al. 2006). After alignment, the datasets were manually edited in MEGA. Phylogenetic analyses of DNA (or protein) sequences are important tools for studying the evolutionary history of different organisms (Nei & Kumar 2000, Salemi &Vandamme 2003). The true phylogenetic tree is almost always unknown. None of the tree-building methods is perfect or superior to others, and different data sets seem to favor different algorithms. Therefore, it is advisable to employ more than one method for each data set, which practice we applied in this study. In studies I-IV, a combination of distance methods (neighbor-joining analysis), parsimony methods (maximum parsimony), likelihood methods (maximum likelihood) and Bayesian inference were used. In studies I-II, three different approaches for phylogenetic analyses were employed. A neighbor-joining analysis (NJ) with the Kimura 2-parameter (K80) substitution model switched on and a maximum parsimony (MP) analysis were performed using MEGA and Bayesian inference (BI) with MrBayes v3.1.2 (Ronquist & Huelsenbeck 2003). The reliability of each interior branch of the tree was examined using a bootstrap test (Felsenstein 1985). Phylograms presented in studies I-II were obtained from the NJ analyses. In studies III-IV, a maximum likelihood analysis (ML) was performed using RAxML 7.0.4 (Stamakis et al. 2008), run on the CIPRES Portal at the San Diego Supercomputing Center (Miller et al. 2009), and a MP analysis using TNT v1.1 (Goloboff et al. 2008) was run on the computer clusters of the CSC, ITS Center for Science, Espoo, Finland, and BI with MrBayes v3.1.2. Phylograms presented in studies III-IV were obtained from the ML analyses. The reliability of each interior branch of the tree was examined using a bootstrap test (Felsenstein 1985). In the TNT parsimony analysis, gaps were coded as a fifth state (using gaps as information). In parsimony analyses, a fifth state coding has been reported to recover a more accurate tree reconstruction compared to t
