On multiplicities of non-isolated intersection components (Q1187171)

In Curves Semin. at Queen's, Vol. 2, Kingston/Can. 1981-82, Queen's Pap. Pure Appl. Math. 61, Exposé A (1982; Zbl 0599.14003), the author and \textit{J. Stückrad} constructed an intersection theory in the projective space \(\mathbb{P}^ n_ k\) \((k\) an algebraically closed field) as follows: Let \(X,Y\) be pure dimensional subschemes of \(\mathbb{P}^ n_ k\). Then there exists a collection \({\mathcal C}\) of irreducible subvarieties of \(X\cap Y\) and intersection numbers \(j(X,Y;C)\geq 1\) for each \(C\in{\mathcal C}\) given by lengths of certain primary ideals such that \(\deg(X)\deg(Y)=\sum_{C\in{\mathcal C}}j(X,Y;C)\deg(C)\). In the paper under review the author investigates the question as to which components of \(X\cap Y\) (especially nonisolated ones) contribute to this intersection theory, i.e. belong to \({\mathcal C}\). Let \(X_ i,Y_ j\) be the irreducible components of \(X,Y\), respectively. The answers depend upon the relationship between the integers \(\dim(X\cap Y)\), \(\dim(X_ i\cap Y_ j)\), \(e\) and \(r_{ij}\), where \[ e=\dim(X\cap Y)-\dim(X)- \dim(Y)+n \] and \(r_{ij}\) is the largest integer such that \((X_ i\cup Y_ j)_{\text{red}}\) is contained in an \((n-r_{ij})\)-plane. In particular, all irreducible components of \(X_ i\cap Y_ j\) do belong to \({\mathcal C}\).