Molecular technology has shown the classical, morphologically defined groupings of sequestrate cortinarioid fungi to be artificial and in need of revision. However, these same molecular studies have highlighted morphological characters, such as spore shape and ornamentation, that have proved useful for distinguishing phylogenetically informative groups. This observation underpins the hypothesis of this study: that the numeric analysis of selected morphological characters can provide the same picture of the diversity of, and relationships among, sequestrate cortinarioid fungi as that recovered from phylogenetic analysis of rDNA sequence data. Sequestrate fungi are those in which the spores mature inside an enclosed fruit body, remaining there until the fruit body decomposes or is eaten. For the purposes of this thesis the following genera are considered to contain cortinarioid sequestrate fungi: Auritella, Cortinarius, Dermocybe, Descomyces, Hymenogaster, Hysterogaster, Inocybe, Protoglossum, Quadrispora, Setchelliogaster and Timgrovea. This thesis focussed on Australian representatives of these fungi to address the hypothesis outlined above. Four analysis methods were applied to each of three datasets (morphological, rDNA and combined data) in a comparative approach to test the stated hypothesis. The four analysis methods were two multivariate methods: cluster analysis and ordination (by principal coordinates analysis), and two phylogenetic methods: maximum parsimony and Bayesian analysis. Low bootstrap support and Bayesian partition probabilities for phylogenetic analyses of the morphological data indicated this dataset had little to no phylogenetic signal discernable by parsimony and Bayesian analyses. Different analyses of the morphological data differed in the way they grouped the collections. The type of clustering method used affected the pattern of relationships recovered. The coding of the data had a much more substantial effect on the patterns of relatedness suggested by the multivariate analyses. Despite the low level of phylogenetic information and agreement between analyses of the morphological data it was found that some collections were consistently grouped together. This included the separation of the Cortinarius-like collections from the Descolea-like collections and the relatively consistent grouping of some pairs of collections and some larger groups. Thus, despite the limited phylogenetic signal of the small morphological dataset and the artefacts of coding, some relatively consistent groups were recovered. Separate analyses of the Cortinarius-, Descolea- and Hebeloma-like ITS sequences recovered similar patterns to published phylogenies. The inclusion of more sequestrate taxa and a greater sample of Australian collections than previous studies, indicated that both Timgrovea subgenera nest among the Descolea-like collections and that hitherto undiscovered lineages of Descolea-like fungi are represented among the collections in Australian herbaria. The Cortinarius-like fungi fall within clades recognised by published phylogenies. Similar topologies were supported by both Parsimony bootstrap and Bayesian partition probability values for analyses of the molecular data including the separation of Cortinarius-like collections from Descolea-like collections. However neither of these methods of analysis and evaluation yielded well-resolved deeper nodes for either of these two major clades. Comparable clades/clusters of Cortinarius- like and Descolea-like collections were found in all analyses of the molecular data. Thus phylogenetically distinct groups of cortinarioid sequestrate fungi could be consistently distinguished using ITS molecular data, but not confidently related to one another. The ratio of molecular to morphological characters (741:16) meant the patterns observed for the combined analyses were more similar to those observed in analyses of the molecular data than those of the morphological data. This included the recovery of substantially similar clades/clusters to those recovered by analyses of the molecular data alone. The value of combining the morphological and molecular data as analysed is questioned despite the congruence of the datasets according to the Incongruence-Length Difference test. Differences between the molecular and combined datasets arose primarily where the molecular data grouped collections that were also grouped by the morphological data. The numeric analysis of the selected morphological characters as carried out in this study did not recover the same pattern of groups and relationships among the cortinarioid sequestrate fungi as phylogenetic analyses of ITS data. The composition of groups recovered using the morphological data alone or as part of the combined dataset, and the relationships between those groups, differed from those recovered from the molecular data alone; although there were similarities between groups recovered from different datasets. The ability of this thesis to conclusively address its fundamental hypothesis was compromised by limitations of the study such as taxon sampling, character selection, character coding and the poor resolution of the ITS phylogeny. Acknowledging these limitations, and that some similar groups were recovered, the results of this thesis do not support its stated hypothesis that the numeric analysis of selected morphological characters can provide the same picture of the diversity of, and relationships among, sequestrate cortinarioid fungi as recovered from phylogenetic analysis of rDNA sequence data.

To improve fungal identification using a single DNA sequence, I introduce the Barcode Similarity Group (BSG) defined as a cluster of sequences that share greater than or equal to a threshold amount of genetic similarity with each other. As a test case, I created 393 BSGs from 2463 Cortinarius ITS sequences using a 94% similarity cut-off value in DOTUR. Some BSGs may contain multiple species. The BSG database was used to label environmental sequences, find misidentified or mislabeled sequences, and find potential cryptic species and novel species. Expert taxonomists will be needed to perform detailed morphological and phylogenetic studies to identify the individual species within each BSG. The main advantage of using BSGs is that it clusters together sequences using total genetic relatedness and does not rely on any taxonomy for identification. A website was created where the RDP Classifier is used to classify a query sequence into a BSG.

To improve fungal identification using a single DNA sequence, I introduce the Barcode Similarity Group (BSG) defined as a cluster of sequences that share greater than or equal to a threshold amount of genetic similarity with each other. As a test case, I created 393 BSGs from 2463 Cortinarius ITS sequences using a 94% similarity cut-off value in DOTUR. Some BSGs may contain multiple species. The BSG database was used to label environmental sequences, find misidentified or mislabeled sequences, and find potential cryptic species and novel species. Expert taxonomists will be needed to perform detailed morphological and phylogenetic studies to identify the individual species within each BSG. The main advantage of using BSGs is that it clusters together sequences using total genetic relatedness and does not rely on any taxonomy for identification. A website was created where the RDP Classifier is used to classify a query sequence into a BSG.