Presentation of names.

Names of taxa at all ranks are italicised following Thines et al. (2020).

Specimen selection.

Fungarium specimens collected in Australia and identified as Camarophyllopsis or Hodophilus in collections were selected for examination from the National Herbarium of Victoria (MEL), the Queensland Herbarium (BRI), the NSW Plant Pathology & Mycology Herbarium (DAR), the Tasmanian Herbarium (HO), the Western Australian Herbarium (PERTH), the University of Tennessee Fungal Herbarium (TENN-F) and the John T. Waterhouse Herbarium (UNSW). Additional specimens from MEL that were identified in collections as Marasmius or indetermined Fungi were included for examination because their morphology was consistent with Hodophilus.

Morphological examination.

The macromorphological description for the novel species is assembled from descriptive notes and photographs of fresh collections recorded by collectors, as well as examination of dried material. Colours of fresh collections are described based on field notes made by collectors in daylight conditions and, where possible, according to Kornerup and Wanscher (1978). Key characters of existing species are a precis of descriptions from Young (2005).

The micromorphological description for the novel species is based on examination of dried fungarium specimens, from which hand cut sections were rehydrated in 5% potassium hydroxide (KOH) following Tang et al. (2023) and then stained with Congo red to easily observe cell walls. Microscopic features were observed and photographed using an Olympus BX51 microscope (Olympus, Tokyo, Japan) with differential interference contrast and an Olympus DP73 camera attachment. Measurements were taken at ×400 or ×1000 (with oil immersion) using measurement tools in Olympus cellSens standard (v. 1.16). Basidiospore measurements were recorded from at least 10 spores per specimen examined. Basidiospore measurements are provided in descriptions as the observed range of values, then as the range of means per specimens examined with the grand mean italicised. The number of spores measured from the the number of specimens examined is provided after the measurements (e.g., n/x), where n indicates the number of spores measured and x indicates the number of specimens examined. Measurements of basidia, cystidia, terminal and subterminal elements of the pileipellis were recorded primarily from the holotype and supplemented from additional specimens examined. These are provided as the observed range of values in the description.

DNA isolation, PCR and sequencing.

DNA was isolated from selected specimens (Table 1) using a modified CTAB method summarised in Craig et al. (2023), based on that of Gardes and Bruns (1993). Polymerase chain reactions (PCR) were performed to amplify the internal transcribed spacer (ITS) rDNA and domains D1–D2 of the large subunit (LSU) rDNA in 20 μl reactions using MyTaq Red Mix (Bioline/Meridian Bioscience, London, UK), 1 μl of DNA template and the primers pairs ITS5/ITS4 or ITS1F/ITS4 (White et al. 1990, Gardes & Bruns 1993) and LR5/LROR (Vilgalys & Hester 1990, Vilgalys Lab 1992), respectively, using protocols outlined by Craig et al. (2023). Sequencing was undertaken by the Australian Genome Research Facility (AGRF, Melbourne, Australia) and chromatograms were aligned, manually checked and edited using Geneious Prime 2021.0.3 (https://www.geneious.com) to generate consensus sequences (Table 1).

Phylogenetic analyses.

ITS and LSU sequences generated in this study were assembled with sequences from species of Hodophilus and the closely related genus Clavaria Vaill. ex L. that were downloaded from NCBI GenBank, based on the phylogeny of Birkebak et al. (2013, 2016), Adamčík et al. (2016, 2017, 2020) and Tang et al. (2023). Additional Hodophilus sequences were downloaded from UNITE (Abarenkov et al. 2023) by searching for sequences (search by NCBI+UNITE sequences; filter for Hodophilus) and using MassBLASTer through the PlutoF workbench (Abarenkov et al. 2010) to find sequences from NCBI and UNITE with high similarity to sequences generated from Australian specimens (Supplementary material 1: Table of all sequences). The ITS and LSU datasets were aligned separately using MUSCLE Alignment (Version 3.8.425) (Edgar 2004) implemented in Geneious Prime (1000 maximum iterations, 100 maximum trees, all other settings default), followed by manual editing including trimming of sequence ends. Preliminary maximum likelihood phylogenetic analyses were conducted separately for the ITS and LSU alignments to check for incongruence. We did not detect incongruence, so we concatenated the ITS and LSU alignments and then manually partitioned for ITS1–5.8S–ITS2–LSU.

Maximum-likelihood (ML) phylogenetic analyses were performed for the ITS1–5.8S–ITS2–LSU alignment using command-line IQ-TREE 2.2.2.6 (Nguyen et al. 2015, Minh et al. 2020). IQ-TREE implemented partition models (Chernomor et al. 2016) to test merging partitions and ModelFinder (Kalyaanamoorthy et al. 2017) to test for the best-fit substitution models for partitions according to Bayesian Information Criterion (BIC). Partition models merged ITS1–ITS2 and 5.8S–LSU and ModelFinder selected TPM2u+F+G4 as best-fit model for ITS1–ITS2 and TIM2+F+I+G4 for 5.8S–LSU. The consensus tree was estimated using UFBoot2 (Hoang et al. 2018) to calculate ultrafast bootstrap support (UFBS) values from 10,000 replicates (Supplementary material 4: ML consensus tree).

Bayesian analyses (BI) were performed for the ITS1–5.8S–ITS2–LSU alignment using command line MrBayes 3.2.6 (Huelsenbeck & Ronquist 2001, Ronquist et al. 2012). The GTR+G+I nucleotide substitution model was applied. Two simultaneous runs of two million MCMC iterations were performed, with four heated chains, a heated chain temperature of 0.2, a sampling frequency of 1000, and burn-in length of 25%, which resulted in stable average standard deviation of split frequencies below 0.01. Sequences of Clavaria pullei were included as the outgroup for ML and BI according to the phylogeny of Birkebak et al. (2013, 2016).

As the topology of ML and BI analyses were consistent, only ML UFBS values ≥ 95% and BI posterior probability (PP) ≥ 0.90 are shown at nodes. For clades with ML UFBS values ≥ 95% and BI PP < 0.90, BI PP are shown in parentheses, and vice versa.

Supplementary materials, including the complete list of sequences (Supplementary material 1: Table of all sequences), the concatenated alignment (Supplementary material 2: ITS+LSU alignment), the partition file (Supplementary material 3: ML partition file), the consensus maximum likelihood phylogeny (Supplementary material 4: ML consensus tree) and the data block used for BI analyses (Supplementary material 5: BI data block) are deposited on figshare (https://doi.org/10.6084/m9.figshare.31064017).

[DOI not yet active. Private link for reviewers: https://figshare.com/s/d63cda3b88f742c5990d].