Acanthodians are a poorly understood paraphyletic grade of extinct Palaeozoic fishes. They play an increasingly prominent role in our understanding of vertebrate evolution as part of the chondrichthyan stem-group even though their evolutionary history is scarce. The limited preservation of their mostly cartilaginous skeleton largely results in a bias towards isolated remains such as fin spines and scales. Here, we quantify the quality of the acanthodian fossil record by using a variation of the Skeletal Completeness Metric (SCM), an approach that calculates how complete the skeletons of individuals are compared to their theoretical complete skeleton. A novel Soft Tissue Completeness Metric (STCM) is introduced to estimate the percentage of soft body tissue preserved as an alternate measurement of completeness. Completeness scores for >1600 specimens comprising >300 taxa obtained from museum collection visits and literature surveys were assembled into a database. Acanthodian completeness peaks in the Lochkovian, Givetian, Bashkirian-Moscovian, and the Asselian, with lowest scores in the Llandovery. There is no correlation between acanthodian taxonomic richness and completeness. Acanthodians show a significantly lower completeness distribution than many tetrapod groups, but a similarly low distribution to bats. Skeletons deposited in freshwater are significantly more complete than in marine environments where sea level significantly negatively correlates with observed completeness. Our assessment of completeness reveals only weak spatial biases influencing the acanthodian fossil record while environmental biases are much higher. Variation in completeness will have an impact on the utility of characters for assessing the phylogenetic relationships and inclusion of the various acanthodian groups.

All information on data collection as well as specific details can be found in the Material and Methods section of the manuscript. 

The summary and input sheet section of the excel file has been formatted specifically to be load into the R software as easily as possible.