Data acquisition
Records of Cenozoic fossil fungi were acquired from both online peer-reviewed literature and literature from the John Williams Index of Palaeopalynology (JWIP) in the Natural History Museum, London. Miocene-dated fossil records were prioritised for transcription. From these sources, information about the location, geologic layers, and fossils present were recorded (Fig. 2). Location coordinates were either transcribed directly from the literature or if not available, were estimated using the description of the location in the source. In the latter case, this estimation was flagged in the ‘Notes’ column of the dataset.
Methodology flowchart showing the synthesis of fossil fungi data into the MyCeno 2.0 dataset. Whilst information was extracted verbatim from online literature and literature from the John Williams Index of Palaeopalynology (JWIP) in step 1, steps 2 to 5 involved reviewing and updating of the geologic age, fungal nomenclature, and taxonomy using secondary literature, Index Fungorum, the Catalogue of Life (COL), and the Global Biodiversity Information Facility (GBIF).
Updating age estimation and grading dating uncertainty
Age estimations were reviewed and updated, where possible, for obsolete or wide-ranging estimations. Newer scientific literature was consulted to check whether formations had had their ages updated. These updates are flagged under ‘Notes’, and the secondary literature sources which informed the new age estimates are referenced under the ‘Supporting_literature_IDs’ column of the dataset.
The level of dating uncertainty was determined for all records using a qualitative grading system, similar to Pound & Salzmann15, to be able to compare reliability of dating techniques across the dataset15 (Table 1). A grade of 1 indicates a very reliable age estimation based on robust dating evidence including radiometric or isotopic dating. Grades 2–4 were given if magnetostratigraphy or biostratigraphy dating techniques were used (4 for a wide numerical range and limited evidence, 2 for a narrow numerical range and multiple taxa used for biostratigraphy). Grade 5 was given if there were vague or no dating techniques mentioned in the source, or just lithostratigraphy was used. Dating techniques and dating certainty grades were based on the new updated age estimates from step 2 of the methodology workflow (Fig. 2).
Nomenclatural and taxonomic review
Correct scientific names, synonyms, and higher taxonomic classifications for the fossils were determined using the Index Fungorum and Catalogue of Life databases16,17. On a few occasions, when names were not in the IF database, The Kalgutkar and Jansonius Database of Fossil Fungi was consulted12. For 15 fossil types (491 fossil records; 14.1% of records), no correct name was found for the fossil names given in sources. This includes those with descriptive titles only (e.g. hyphae, fungal spores, fungal remains etc.). For these kinds of fossils although ‘Correct_name’, ‘Correct_name_author’, ‘Is_genus_extant’, and ‘Higher_classification’ entries were left empty, recording their occurrence provides direction for future re-examination and re-collection efforts.
To determine whether a fossil genus is extant, the Global Biodiversity Information Facility (GBIF) and scientific literature were consulted18. GBIF was used to check for images of occurrences and whether the occurrences are annotated as fossil or living specimens. Scientific literature was consulted to see if living occurrences might be missing from GBIF.
If the fossil taxon is considered extant, this taxon was used for the nearest living relative (NLR) of the fossil. Otherwise, the next above taxonomic classification was checked for whether that was extant or not. The lowest possible extant taxonomic classification was used as the NLR for a given fossil taxon. NLRs were never assigned based on speculative comments of morphological affinity between fossil and modern fungi taxa.
