Vanishing cycles for analytic maps (Q3973662)

[For the entire collection see Zbl 0723.00028.] The author proves: Let \(f_ 0:\;(\mathbb{C}^ n,0)\to(\mathbb{C}^{n+1},0)\) be a holomorphic map germ which is finitely \({\mathcal A}\)-determined and let \(f_ t:\;B_{\epsilon}\to\mathbb{C}^{n+1}\) be a good representative of a stable perturbation of \(f_ 0\). (1) Assume \((n,n+1)\) in the range of nice dimensions (i.e. \(n\leq 6\)), then the image \(X_ t\) of \(f_ t\) has the homotopy type of a bouquet of spheres. (2) If \(n=2\), then the number \(\sigma\) of spheres in the bouquet satisfies: If \(\text{cod}({\mathcal A}_ e,f_ 0)\) denotes the dimension of the parameter space of the universal unfolding of \(f_ 0\), then \(\text{cod}({\mathcal A}_ e,f_ 0)\leq\sigma\) with equality if \(f_ 0\) is weighted homogeneous. There is a remarkable analogy of this result to the case of complete intersection map germs \((\mathbb{C}^{n+k},0)\to(\mathbb{C}^ k,0)\): If we replace in (1) ``image'' by ``fibre'' and ``\(\mathcal A\)'' by ``\(\mathcal K\)'' we obtain well-known statements about the Milnor fibre. If we replace in (2) ``\({\mathcal A}\)'' by ``\({\mathcal K}\)'' and ``image'' by ``discriminant'' we obtain the known results relating the Tjurina and Milnor number. It is a challenging problem to find a general theorem which includes these as special cases.