[en] A discussion of some aspects of the U(1) problem is given and the large N point of view is considered. The necessity for a vector ghost gluonic particle, in order to resolve the U(1) problem, is stressed. Scalar ghosts are shown to be insufficient. An analysis of the eta-eta' mixing problem is made and it is suggested that the singlet decay constant may be nearly twice as large as the other octet decay constants. PCAC corrections are then considered and the above conclusion is reaffirmed. Quark mass values are given and it is pointed out that in the usual scheme, < anti ss> approximately equal to 1.48 < anti uu>. The compatibility of the above with the eta→3π decays is also checked. (author)