1. Concerning the interaction analysis, the present results are obtained from a linearized approach that was first used in the early 1950's. The results are also derivable, however, from the viewpoint of geometric acoustics by utilizing an energy density invariant for acoustic waves pr,opagating in a nonuniform moving medium. This invariant was f i r s t derived by Blokhintsev 34 and has been used to describe sonic born pr 0pagation 3 (the sonic boom is a weak shock r!ave system). The wavefronts are found to be focused in the portion of the boundary layer. Borrowing wavepropagation terminology, this focal region is herein termed the "caustic region." Examples of wave behavior a t a caustic are cited and estimates of the amount of focusing are presented.
2. The most thorough analytical study to date of the interaction between the noise and the laminar boundary layer in supersonic wind tunnels is that of Mack.36,37 His calculations were for frequencies in the range here Kendall's free stream measurements" shoned significant energy content. One finds, however, that the wavelengths considered were long in comparison to the laminar boundary layer thickness (10 6 - 350 6). Thus Mack's results can be considered to describe the interaction with the lower frequency components of the noise field, i:hile the present results describe the interaction with the weak shock fronts, or, in the sense of Fourier decomposition, the very highfrequency components of the field (frequency -).
3. Inasmuch as the detailed noise measurements from which one can confidently extract a noise field model are not yet available, one must turn to the source of the noise. the turbulent boundary layer, to obtain further guidance. The literature on turbulent shear flow is immense and growing, and the prevailing views on the subject have been undergoing considerable change. In the mid-1950's the older view of turbulence as a mean flow plus random fluctuations began to give ground to a more mechanistic view. This was the result of increasing recognition that there appeared to be some order or coherent motion in turbulent flows. The trend in experimental research then shifted to the revelation and exploration of ordered or quasiordered motions, and within the last 10 years. computers, and optical methods, much information has been acquired about the structure of the turbulent boundary layer. Reviews of this research may be found in References 39-42. Some of the latest findings are cont ined in the coherent structure workshop