Abstracts Numbering of atoms in relatively large molecules, such as fullerenes appears for most part to be arbitrary or based on ad hoc schemes. We argue in favor of the use of a particular canonical labeling of atoms in molecules based on the smallest possible binary molecular code obtained from the adjacency matrix when its rows are read from left to right and from top to bottom. The approach has been illustrated with buckminsterfullerene. We have outlined advantages of the approach and have shown that finding canonical labels is practical even in the case of large regular graphs.
