The question of cognomina tends to be over-simplified in the elementary textbooks, where one meets such statements as: ‘Most Romans had three names … of which the third defines the familia within the gens.’ Such brevity is doubtless pedagogically convenient, but it has its dangers, particularly in respect of one of the most important periods in Roman history and literature, the late Republican; for there exceptions to this rule are so numerous and important as to make it almost meaningless. To demand greater accuracy in this matter is not mere pedantry; already a tradition is being established powerful enough to influence those whose business it is to know better, even leading, as I shall show later, to the misinterpretation of texts.