In this paper we solve several boundary value problems for the Helmholtz equation on polygonal domains. We observe that when the problems are formulated as the boundary integral equations of potential theory, the solutions are representable by series of appropriately chosen Bessel functions. In addition to being analytically perspicuous, the resulting expressions lend themselves to the construction of accurate and efficient numerical algorithms. The results are illustrated by a number of numerical examples.boundary value problems | potential theory | corners | elliptic partial differential equations | Helmholtz equation I n potential theory, the Helmholtz equation is reduced to an integral equation by representing the solutions as single-layer or double-layer Helmholtz potentials on the boundaries of the regions. By taking the limit of the solutions to the boundary, the densities of these potentials are shown to satisfy Fredholm integral equations of the second kind.When the boundaries of the regions are smooth, the kernels of the integral equations are weakly singular, and the solutions are also smooth. This environment is well understood; the existence and uniqueness of the solutions follow from Fredholm's theory, and the integral equations can be solved numerically using standard tools (see, for example, ref. 1).When the boundaries of the regions have perfectly sharp corners, both the kernels and the solutions of the integral equations are singular. The behavior in the vicinity of corners of the solutions of both the integral equations and the underlying differential equation have been the subject of much study (see refs. 2 and 3 for representative examples), although the differential equation appears to have received more attention than the integral equations. Comprehensive reviews of the literature can be found in (for example) refs. 4 and 5.The leading singular terms in the solutions, in the vicinity of corners, to both the integral and differential equations are known (for example, ref. 2), and there are a number of theorems describing the spaces to which the solutions belong (for example, refs. 6 and 7). In 1979, R. J. Riddell published a heuristic argument for the existence of a certain asymptotic series for the solutions to the integral equations near the corners, but this line of investigation does not appear to have been pursued further (8).In this paper, we provide a detailed description of the behavior of the solutions to the integral equations in the vicinity of corners, in the specific case of polygonal boundaries. We observe that the solutions in the vicinity of corners are representable by certain series of appropriately selected Bessel functions. The analytical results are used to construct highly accurate and efficient numerical algorithms and are demonstrated by a number a numerical examples. This paper is based on several specific analytical observations, which are described in the following section.