The charm of genetics, along with much of its power, lies in its capacity to reveal connections that could never have been anticipated. Pilia et al.(1996) must have been surprised when they uncovered the gene responsible for the overgrowth and tumor-susceptibility syndrome Simpson-Golabi-Behmel syndrome (SGBS). SGBS is an X-linked disorder characterized by pre-and postnatal overgrowth; numerous morphological abnormalities, including congenital heart defects; dysplastic kidneys; vertebral and rib defects; and postaxial hexadactyly. In addition, patients are at high risk for Wilms tumor and neuroblastoma (Neri et al. 1998). SGBS is caused by disruptions of the glypican 3 gene (GPC3), which, remarkably, encodes a proteoglycan of the glypican family. The developmental abnormalities associated with SGBS demonstrate that cell-surface proteoglycans can affect tissue growth and morphogenesis and may serve as a novel class of tumor suppressors. How a proteoglycan affects growth regulation and tumor susceptibility is not immediately obvious, but several laboratories, including my own at the University of Arizona, have found clues at the intersection of cell biology, biochemistry, and genetics. As is often the case, an understanding of a human disease derives in large measure from studies of different organisms and experimental systems. The story that emerges highlights two fundamental features of cell surfaces. First, the plasma membrane is carpeted with sugars, principally as oligosaccharides attached to lipids and proteins. Second, the sugars found on the cell surface show a great deal of structural diversity, with different forms represented in tissue-specific patterns. The function of glycosylation of cell-surface molecules has long been uncertain, but it now appears that one class of sugar-modified proteins,