Canoco: a Fortran program version 2.1 (Ter Braak, 1988) was used to
relate fungal population structure to changes in water variably. Canoco is
particular efficient for ordination of sparse data sets (data containing many zero
values compared to the number of non zero values). The theory of applying
ordination to this type of data isdescribed in Ter Braak (1986). This
multivariate technique embraces canonical correspondence analysis (CCA) that
has been used here. The data are arranged on two axes. The numerical
importance of an axis is judged by looking at its Eigen values. The Eigen
values are a measure of separation of the species distributions along the
ordination axis (Ter Braak, 1987). Theyare always between zero and one.
Higher Eigen values indicate important ordination axis