An important class of conformal array antennas for modern military and commercial communications systems is a single-ring arc array. In arc arrays, the array synthesis must be performed for a partial sector of the entire cylindrical array. Cylindrical arc array pattern synthesis is complicated because the apparent spacing between the elements at various observation angles is not equal nor constant with angle. The element and array factors are not always separable and therefore the pattern function is not a simple polynomial. This precludes the use of the mature linear array synthesis methods that have allowed one to conveniently link polynomial-based distribution functions for linear arrays to their pattern performance. To obtain the distribution functions, a design program based upon genetic algorithm (GA) optimization was developed. The emphasis of this paper is not the development of the best GA, but the utilization of GA in generating the optimal distribution sets to understand the physics behind are arrays. The GA optimizes the array amplitude and phase coefficients, given the array radius, the total number of elements around the entire ring, and the number of exited elements.