A note on imbedded prime divisors (Q2042123)

The authors give some counter examples showing that the following result proved by [\textit{L. J. Ratliff jun.}, Proc. Am. Math. Soc. 101, 395--402 (1987; Zbl 0643.13001), Remark 2.2] is not correct: If \(R\subset T\) are Noetherian rings such that there does not exist any integrally dependent adjacent Noetherian rings between them, then for each \(\bar{c}/\bar{b}\in T /Z\) (where \(Z = \mathrm{Rad}(T ) = \mathrm{Rad}(R)\) and \(\bar{b}, \bar{c}\) regular in \(R/Z\)), we have either \(\bar{c}/\bar{b}\in R/Z\) or \(\bar{b}/\bar{c}\in R/Z\), and so \((R/Z)[\bar{c}/\bar{b}]\) is a localization of \(R/Z\). Moreover, some additional conditions are given under which this result holds. In particular, when \(R\) and \(T\) are integral domains, then this result holds if and only if \(R\) and \(T\) are local.