States that condensation on windows is a major issue for building owners and managers. Tests were devised to counter the problems of sealing windows, test temperatures and cost and simulation testing for conventional testing procedures, and a database of windows that have undergone both tests and simulations for resistance to condensation was compiled. Three double hung windows of varying materials were tested as well as designs for air infiltration at several levels of airtightness. Air leakage rate for each window was tested by progressively unsealing it, then unlocking it, and finally by opening it slightly. Goes on to describe test procedure, simulation procedure, air infiltration test results, condensation potential simulation results. Concludes that testing sealed or unsealed windows in a pressure chamber makes little difference to surface temperature; for the windows tested, test and simulation results are close for glass-edge values; when experimenters correct for three-dimensional vertical convective motion in hollow-frame cavities, test and simulation values are also fairly close. Recommendations for testing and additional research were: seal windows before testing for condensation potential to prevent uncertainties associated with air movement through windows; sealing should have little impact on results; conduct condensation-potential tests at -18 Deg. C, and you can do U-factor testing simultaneously; researchers must develop a model for three-dimensional convective motion in frame cavities. Until such a model exists the correction used appears to provide reasonable estimates of frame surface temperatures.