In this thesis phytocoena and mycocoena of ectomycorrhizal fungi and saprotrophic fungi in roadside verges planted with trees are described independently. An attempt is made to indicate which environmental variables are most important in the distinguished communities. Parasitic fungi on trees, arthropods and other fungi and saprotrophic lignicolous species were not included in community definition.In roadside verges in the phytogeographical Drenthian district 76 plots were selected, 53 planted with Common Oak (Quercus robur; "Oak plots") and 23 with Beech ( Fagus sylvatica ; "Beech plots").12 beech plots were situated in an open landscape ("open") and 11 along roads inside forests ("shady"). These plots varied widely with regard to soil fertility.The oak plots were divided into 34 open plots, 10 shady plots and 9 "half-open" plots, i.e. bordered at one side by forest. In the open plots three age-classes were distinguished: 5 plots with young trees (up to 20 years old), 12 plots with middle aged trees (20 to 50 years old) and 17 with old trees (more than 50 years old). In the remaining plots old trees were present. In oak plots, too, there is a large variation regarding the soil fertility. Plots were always 100 in long, irrespective of the width of the verge and were selected on the basis of a sufficient lengthwise homogeneity of the phanerogam vegetation. A large number of environmental variables was measured.Vegetation relevés were made in 1987 according to the Braun-Blanquet method. For mycological purposes, the plots were visited during the autumns of 1986, '87 and '88 once every 3 to 4 weeks. All fungi were counted and identified.The data were processed with the aid of the computer programmes TWINSPAN for vegetation classifications and CANOCO for ordinations and correlations with environmental variables.In the beech plots 134 species of green plants, 105 species of ectomycorrhizal and 153 species of saprotrophic fungi were found.On the basis of the green plants two vegetation types could be recognized, one with open and one with shady plots. The former was divided into two subtypes, one with a poor and one with a semiruderal vegetation. Using the ectomycorrhizal fungi the plots were divided into a species poor and a species rich type with two subtypes. The former did not correspond with a vegetation type but the latter two subtypes corresponded to a limited extent with vegetation types. The two communities that were recognized among the saprotrophic fungi corresponded well with the vegetation types. The better correspondence of communities of vascular plants with those of saprotrophs than with those of ectomycorrhizal fungi indicates that plants and saprotrophs react more in the same way to the environmental factors than the ectomycorrhizal fungi.Environmental factors important for plants and saprotrophs are exposition of the plot, thickness of the organic layer, sodium concentration and Ellenberg N-indication values. Important for ectomycorrhizal fungi are: higher nitrate, potassium and magnesium concentrations for the species poor type and "openness", thickness of the organic layer and age of the trees for the other types.In the oak plots 198 green plant species, 144 ectomycorrhizal and 214 saprotrophic species were found.Three main vegetation types were distinguished, the Mnium hornum type of forest roads, the semiruderal Anthriscus sylvestris type of open to shady plots and the Hypochaerisradicata type of open, rather nutrient poor plots. For the ectomycorrhizal fungi, four types were recognized: the Xerocomus rubellus-, the Russula ochroleuca-, the Cortinarius erythrinus- and the Hebeloma mesophaeum type, characteristic for shady plots, semiruderal open to shady plots, open nutrient poor plots and open plots with young trees respectively. In the ordination, the environmental factors tree age, exposition of the plot and Ellenberg N indication values were most important. In the plots of the R. ochroleuca type a thicker organic layer and larger amounts of soluble and total nitrogen were present.Based on the saprotrophic fungi, three communities were distinguished: the Psathyrella fulvescens-, the Mycena avenacea- and the Collybia cookei type. The latter is a small, weakly characterized type. The P.fulvescens type comprises shady to open plots with several vegetation types, the M.avenacea type open plots with a short, grassy vegetation. Most important environmental factors for the distinction of the types were "openness" tree age, Ellenberg N- indication values and thickness of the organic layer. The communities of saprotrophic fungi corresponded better with the vegetation types than the communities of mycorrhizal fungi.The classification of the oak and beech plots together on the basis of the green plants is largely analogous to the classifications of oak and beech plots separately. For the saprotrophic fungi, Ellenberg N values, thickness of the organic layer and the openness of the plots are determining factors in the classification of the plots. However, for the classification of the plots using the ectomycorrhizal fungi the tree species is the most important parameter. Within the oak and beech group, the classification resembles the classifications of oak and beech plots separately. In plots with eutraphent vegetation nonhost-specific ectomycorrhizal fungi dominate. In such plots the ectomycorrhizal mycoflora of oak and beech plots is more or less similar.A comparison of the fungal communities of roadside verges with forest communities of the same tree species was made. Comparison with Dutch oak forest communities revealed that 42 ectomycorrhizal species were found to be differential for roadside verges and 39 for forests. 16 species were indifferent. Among the saprophytes, 24 species were differential for roadside verges, at least 62 for forests and 24 were indifferent. Only the Dicrano-Quercetum showed resemblance with some types of oak plots. The main difference with the other types of oak forest is the larger number of ectomycorrhizal fungi in roadside verges. Regarding the saprotrophic fungi, roadside verges differ profoundly from forests. Terrestrial raw humus inhabiting and lignicolous fungi are mostly restricted to forests, whereas typical grassland fungi were mostly found in roadside plots.Dutch oak forest communities showed more similarity with the roadside verge communities than those from other parts of Europe. Generally, communities with little or no organic layers on nutrient-poor soils have a large proportion of ectomycorrhizal species in common with the analogue roadside communities.A classification is presented for ectomycorrhizal roadside fungi based on their assumed restricted occurrence in the roadside habitat. In roadside verges a number of threatened fungi occur that presumably have disappeared from forests with the same tree species.In 25 plots of the Hypochaeris radicata-Quercus type with trees of three different ages successional series were studied. The number of ectomycorrhizal species increases with the tree age. This is in contrast to data from forests in the literature, where the species number decreases after an initial increase. The differences in soil and the management practices in roadside verges may explain this discrepancy. Eutrophication and litter accumulation cause a decrease in the species number, resulting in low species numbers in eutrophicated places and/or in thick litter layers, even under old trees.A new, preliminary classification of ectomycorrhizal fungi regarding their respective optima during the sucession of the site, based on the data from roadside verges is presented.In a homogeneous roadside verge with 100 year old Common Oaks along a canal the effects of various management treatments were studied. During the years 1987-'91, 5 different treatments were applied: a. mowing without removal of the hay, b. mowing with removal of the hay, c. nonrecurrent removal of sods in combination with mowing without removal of the hay, d. N-fertilization during the first 3 years in combination with mowing without removal of the hay, e. no treatment. Treatment "a" served as control. Each treatment was carried out sixfold in plots of 50x4.5 m 2. The phanerogam vegetation, some soil properties and ectomycorrhizal samples from treatments b, c and d were analyzed as well.Removal of sods caused an immediate decrease of sporocarps of mycorrhizal fungi but after 3 years the species numbers had largely recovered. However, there was some shift in species composition. After fertilization a strong decrease of ectomycorrhizal fungi occurred. The saprotrophs were significantly afflicted by removal of sods. Fertilization resulted in a decline in saprotroph species numbers but the production did not change significantly. Treatments b and e showed no significant differences with the control. The soil chemistry was not influenced by the treatments except for higher concentrations of N in fertilized plots.In eight open beech plots, viz. four in plots rich in ectomycorrhizal species on nutrient-poor soil and four in species-poor and nutrient-rich soil, root samples were taken. In the field-experiment root samples were taken two years after the start of the treatments b, c and d.In the beech plots a positive correlation was found between the number of mycorrhizal root tips and the number of ectomycorrhizal species found during the fieldwork. However, there was no correlation with the biomass production of the sporocarps. A non-significant negative correlation was found between tree vitality and the number of mycorrhizal root tips. No significant correlations at all were found between the treatments mowing with removal of the hay, removal of sods and fertilization regarding: total root length, number of mycorrhizal root tips, degree of ramification, percentage of mycorrhizal root tips and number of types of mycorrhizal root tips. In fertilized plots the samples contained more mycorrhizal root tips with a relatively large proportion of Cenococcum.In this study, descriptions, drawings and observations are presented of rare, critical or less well-known macromycetes that were encountered in the oak plots, the beech plots, and in the experimental site. Special attention is paid to the genera Cortinarius S.F. Gray emend. Fr., Hebeloma (Fr.) Kumm. and Russula Pers.. Psathyrella rhombispora Keizer & Arnolds is presented as a new species. Russula cicatricata Romagn., Russula elaeodes (Bres.)Romagn. and Russula purpurata Crawsh. are reduced to formae of Russula graveolens Romell in Britz..