The 140 000 or so fungal species reported so far are heterogeneously defined based on varying criteria such as morphological, physiological, mating, and (or) molecular features. Incongruences are common among traits used to separating closely related species and it is often difficult to compare fungal taxonomic groups defined based on different species recognition criteria. Though DNA sequence-based classification and identification have been made, a consensus has not been reached, primarily due to intrinsic limitations in the proposed one or a few genes. Here, I argue that the fundamental reason for the observed inconsistencies is that speciation is a stochastic process with the emergence and fixation of different traits influenced differently by many non-deterministic factors such as population size, random mutation, mode(s) of reproduction, selection imposed by interacting biotic and abiotic factors, and chance events. Each species concept attempts to capture one or a few traits emerged in the continuous process of speciation. I propose that a genome sequence-based classification and identification system could unify and stabilize fungal taxonomy and help integrate taxonomy with other fields of fungal biology. The genomic species concept could be similarly argued for other groups of eukaryotic microbes as well as for plants and animals.