"Predication," writes W. V. Quine, "joins a general term and a singular term to form a sentence that is true or false according as the general term is true or false of the object, if any, to which the singular term refers".1 The view of predication expressed by Quine in the quoted passage is not restricted to Quine; P. F. Strawson, for example, though perhaps not a rabid supporter of Quine's choise of words, is on record as finding the theory congenial with his own views.2 Quine has also written that "so long merely as the predicated general term is true of the object named by the singular term... the substitution of a new singular term that names the same object leaves the predication true."3 Nevertheless, most of my efforts will le directed at establishing that the theory of predication expressed in Quine's words is nonextensional. To be precise about my quite limited objective, I need Quine's help just once more. He writes that "in an opaque construction you also cannot in general supplant a general term by a coextensive term (one true of the same objects)... without disturbing the truth value of the containing sentence. Such a failure is one of the failures of extensionality."4 The theory of predication under consideration is, I claim, non-extensional in the sense that it does not satisfy the extensionality principle that coextensive general terms substitute for each other salva veritate; proof of this claim is my major objective. My secondary objective is to elicit some of the implications of the claim that the theory of predication under discussion is nonextensional.