test of experimental confirmation…. Accordingly, any conclusions as to the accuracy of the entire theory must be postponed until experimental results are available.”

In his doctoral thesis, “Studies of the Validity of the Hydraulic Analogy to Supersonic Flow,” Don investigated experimentally the nature of pressure variations around airfoils at supersonic velocities from measurements of oblique hydraulic jump characteristics resulting from airfoil-shaped obstacles in a supercritical flow. Because there were no adequate high-speed wind tunnels to test his thesis, Don designed and supervised the construction of a unique high-velocity tilting-flume facility, a testimony to his prowess as an engineer.

Upon completion of his doctorate in 1950, Don accepted an appointment as assistant professor of hydraulics at MIT. That year was most significant, though, because he married his beloved companion, Martha Havens, who remained his life partner until his death. On May 2, 2006, Martha died of complications of respiratory disease. She was 82.

Don had an extraordinary career at MIT. His appointment in 1950 coincided with the dedication of a new hydrodynamics laboratory. In 1970, with funding from the founder of the Ralph M. Parsons Company, the laboratory was doubled in size and rededicated as the Ralph M. Parsons Laboratory for Water Resources and Hydrodynamics. Don Harleman followed Arthur Ippen as director of the laboratory from 1973 to 1983.

Don was motivated to solve issues that would improve the quality of life and protect the environment. In pursuit of those goals, he changed research directions several times during his long career. His early work on stratified saltwater systems afforded a natural segue to thermally stratified freshwater systems (lakes and reservoirs). In the mid-1950s, he investigated engineering controls of stratified flow from freshwater reservoirs for the Tennessee Valley Authority. Don and his students established a fundamental model for predicting stratification based on meteorological conditions. The model relied on molecular diffusion for vertical heat transport, a formulation that avoided calibrating an eddy diffusion function to capture turbulent diffusion. As part of a series of studies of stratified reservoirs, Don then worked