A software client usually has few means of assessing the maintainability of a software system in the acquisition process. Software maintenance is often expensive, hence, strategies for assessing the maintainability of complete software systems are important. Assessing the maintainability of complete systems is difficult due to the influence of many different factors such as the people, tasks and tools, in addition to the code. Most research on maintainability focuses on individual classes of individual systems. This paper describes an empirical study where the maintainability of four functionally equivalent systems was assessed using both structural measures and expert assessments. The results suggest that such a combination may be useful. Although, the assessment based on structural measures mostly corresponded with the expert assessments, there were several examples of potential maintainability problems that were not captured by the structural measures.