The widely held assumption about what motivated ''On Denoting'' is irr
econciable with Russell's position shortly beforehand; but discarding
it leaves one with a carefully worked out solution whose problem is mi
ssing. The real motivation is to be found in a notoriously obscure pas
sage in OD, in which Russell exposes a decisive (though easily overloo
ked) flaw in his former theory of denoting; a flaw which also cripples
Frege's theory of sense and reference. A comprehensive account of thi
s passage is the chief concern of the present paper. Recognizing the c
ritical role of this argument of Russell's leads to a more credible ac
count of his argumentation in that essay. It also suggests that the fu
ndamental standpoint underlying The Principles of Mathematics remains
intact. In this light, the appropriation of OD to the philosophy of la
nguage may be misguided.