Using sound to complement graphics and to present data is becoming increasingly popular. Sound can potentially reveal patterns and anomalies in data that are difficult to perceive visually. Moreover, psychological studies show that some types of data are more quickly assimilated when presented with sound; an audio alarm is the classic example. Unfortunately, the lack of standards governing the software interface to sound hardware, as well as the diversity of sound hardware, make it difficult to portably integrate mappings of data to sound ( sonifications ) with existing data visualization tools. This paper describes the design of a sonification toolkit, Porsonify, created to simplify this task.