Glycosylation is the most diverse post-translational protein modification. It is essential for multicellular life and its complete absence is embryonically lethal. Hundreds of specific enzymes are involved in the synthesis of complex oligosaccharide structures that are covalently bound to protein backbones. This process is not template driven and thus results in a huge complexity of glycoproteome, estimated to be several orders of magnitude larger than proteome. Large structural variability provided by glycans represents a significant evolutionary advantage and nearly all proteins invented after the appearance of the multicellular life are glycosylated. Glycosylation represents a way how complex organisms could develop novel structural features without introducing probably deleterious changes in their genome. Intricate mechanisms by which the interplay of gene expression and intracellular localization of their products give rise to specific glycan structures is only starting to be understood, but some evidence suggests that epigenetic regulation of glycosylation might be used to create novel biological structures. Here we suggest a hypothesis that epigenetic regulation of genes involved in glycan synthesis might represent a way how newly developed structural advantages could be transmitted through generations, thus providing a tool for complex organisms to compete with high speed of evolution of unicellular organisms.