(iii) Consistency checks

Having computed various quantities one should of course always double check them with alternative arguments. In this context this is even more important as calculations can be quite lengthy, thus increasing the susceptibility to computational errors. However, besides this ubiquitous reason, one also has to be aware that often some of the assumptions made at the very basis or during the evaluation are not entirely rigorous and an additional verification of the results will at the same time support the reasoning followed. Of course in many cases one may always resort to the perturbation theory in the coupling constant, 1/N, etc., but more in the spirit of the program I will concentrate here on non-perturbative checks.