A software system is an aggregate of communicating modules. The interfaces supporting module (procedure, etc.) communication characterize the system. Therefore, understanding these interfaces (areas of communication) gives a better description of system complexity. Understanding in an empirical sense implies measuring, and measuring interfaces involves examining both the communicational environment and the exchanged data. There are several different measures associated with the communication environment. Obviously, the structure or nesting level at the communication ping is very interesting. The need to measure the data communicated also raises some very interesting questions concerned with data type and expressional form. This paper reports on the efforts at Virginia Tech to measure, and thus capture, the complexities of software interfaces. Analyzing an Ada system of 85,000 lines of code validated the measures proposed here. The results of this research are very encouraging.