A collaboration between UGA scientist Jonathan Amster, Robert Linhardt, at Rensselaer Polytechnic Institute, and Toshihiko Toida, at Chiba University, has led to the first ever sequencing of full length glycans from a naturally occurring proteoglycan. Although much of the structure of the protein portion of bikunin was well known, as was its homogenous N-linked glycan, the O-linked glycan was known to be a chondroitin sulfate that was heterogeneous in length and composition. The CS was released from bikunin, purified into less complex fractions by capillary gel electrophoresis, and the fractions analyzed by Fourier transform mass spectrometry. Composition analysis suggested a small number of chain lengths and degrees of sulfation (less than 20 compositions). Tandem mass spectrometry yielded the surprising result that each composition was a single sequence with a well-defined pattern of sulfation. Even more surprising was the discovery of a conserved pattern of sulfation among all of the components that were analyzed. This provides the first evidence that biosynthesis of glycosaminoglycans is not random, despite its non-template nature.