Density and the circular projection (Q1193060)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Density and the circular projection |
scientific article |
Statements
Density and the circular projection (English)
0 references
27 September 1992
0 references
Let \(D\) be a simply connected domain, not the whole plane, and let \(w=f(z)\) map \(\{| z|<1\}\) one-to-one and conformally onto \(D\). \(R^*\) denotes those accessible boundary points of \(D\) such that \(D\) twists violently about them. For almost every point of the unit circle the finite radial limit \(f(e^{i\theta})\) exists and defines a rectifiably accessible boundary point with complex coordinate \(f(e^{i\theta})\). By a well-known theorem of McMillan, each such accessible boundary point, except for a set of harmonic measure zero, is either a twist point or a point at which \(\partial D\) has an inner tangent. \textit{T. E. Gerasch} [Mich. Math. J. 33, No. 2, 201-207 (1986; Zbl 0605.30020)] showed that except for a set of harmonic measure zero, each point \(w_ 0\) of \(\partial D\) is ``broadly accessible''. This means that there exists a sequence of disks \(D_ n\) in \(D\), of radius \(r_ n\), approaching \(w_ 0\) such that the center of \(D_ n\) can be joined to \(w_ 0\) by an arc of length \(\delta_ n\) and for which \(\lim_{n\to\infty} \delta_ n/r_ n=1\). This can be viewed as meaning that the structure of the boundary is not too bad at most points of \(\partial D\) (with respect to harmonic measure). Here the author proves another result in this direction, this one being a weak version of accessibility of \(w_ 0\) in Stolz angles. This is then applied to prove the following: Let \(L_ \nu\) denote the ray with terminal point \(f(e^{i\theta})\), having inclination \(\nu\), \(0\leq\nu<2\pi\). Let \(m\) denote Lebesgue measure on \(L_ \nu\) and set \[ u(\nu)=\limsup_{r\to 0} {{m((L_ \nu\cap D)\cap(f(e^{i\theta}), f(e^{i\theta})+re^{i\nu}))} \over r}. \] Let \(W^*=\{\alpha\in\mathbb{R}^*\): there exists \(L_{\nu_ i}\), \(i=1,2,3\), at \(f(e^{i\theta})\) such that \(| \nu_ i-\nu_ j|={2\over 3}\pi\), \(1\leq i<j\leq 3\), and \(u(\nu_ i)<1\) for \(i=1,2,3\}\). Then the harmonic measure of \(W^*\) is zero.
0 references
accessible boundary point
0 references
harmonic measure
0 references
Stolz angles
0 references