SUMMARYIn order to assess the factors influencing the extent and composition of ectomycorrhizal infection, and the consequences for seedling growth, seedlings of oak {Quercus robur L,) and birch {Betula pendula Roth,) were established in both field plots and pots of soil from two forested sites (Thetford Chase and Leith Hill), Seedlings were harvested after one growing season, and analysed for ectomycorrhizal infection.When a single root system of pot-grown oak was analysed in detail seven mycorrhizal types were differentiated. The relative abundance of these types varied depending on whether infection was assessed by counting branched mycorrhizas, or individual mycorrhizal tips. This reflected variation in the pattern of branching of the different mycorrhizal types. Of the 41 mycorrhizal types differentiated in the entire survey of 1800 seedlings, five were consistently recorded as unbranched, and six were found to produce second-order branches. Different mycorrhizal types displayed contrasting patterns of abundance, both with respect to site and host species. For example, 21 types were recorded infecting oak, whereas only 15 were found on birch; only three types were found infecting both oak and birch on both sites.In this survey, seedlings tended to be dominated by between one and three mycorrhizal types. Two types were particularly dominant: those formed by Scleroderma citrinum (Pers,) and Paxillus involutus (Batsch) Fr,, which accounted for up to 61 and 92 % of the mycorrhizas of oak and birch respectively. These fungi were exceptional amongst those encountered in the present analysis, in producing both highly branched mycorrhizas and mycelial strands.The relationship between ectomycorrhizal infection and seedling growth was assessed by correlation. Total seedling dry mass was in general found to be positively correlated with the total number of ectomycorrhizal tips, but poorly correlated with the extent of ectomycorrhizal infection, except where seedling growth had previously been shown to be limited by phosphate availability (i,e, in the case of pot-grown birch on soil from Leith Hill), Seedling dry mass was also in general poorly correlated with number of mycorrhizal types, with two exceptions (i,e, field-grown oak and pot-grown birch on soil from Thetford Chase), These results are discussed in the context of the role of ectomycorrhizal infection during seedling establishment, and implications for the methodologies utilized in assessing mycorrhizal infection are highlighted.