Cellobiose dehydrogenase, the only currently known extracellular flavocytochrome, is formed not only by a number of wood-degrading but also by various phytopathogenic fungi. This inducible enzyme participates in early events of lignocellulose degradation, as investigated in several basidiomycete fungi at the transcriptional and translational level. However, its role in the ascomycete fungi is not yet obvious. Comprehensive sequence analysis of CDH-encoding genes and their translational products reveals significant sequence similarities along the entire sequences and also a common domain architecture. All known cellobiose dehydrogenases fall into two related subgroups. Class-I members are represented by sequences from basidiomycetes whereas class-II comprises longer, more complex sequences from ascomycete fungi. Cellobiose dehydrogenase is typically a monomeric protein consisting of two domains joined by a protease-sensitive linker region. Each larger (dehydrogenase) domain is flavin-associated while the smaller (cytochrome) domains are haem-binding. The latter shorter domains are unique sequence motifs for all currently known flavocytochromes. Each cytochrome domain of CDH can bind a single haem b as prosthetic group. The larger dehydrogenase domain belongs to the glucose-methanol-choline (GMC) oxidoreductase superfamily - a widespread flavoprotein evolutionary line. The larger domains can be further divided into a flavin-binding subdomain and a substrate-binding subdomain. In addition, the class-II (but not class-I) proteins can possess a short cellulose-binding module of type 1 at their C-termini. All the cellobiose dehydrogenases oxidise cellobiose, cellodextrins, and lactose to the corresponding lactones using a wide spectrum of different electron acceptors. Their flexible specificity serves as a base for the development of possible biotechnological applications.