The efforts of the Office of Saline Water (U.S. Dept. of the Interior) to develop an economical means of desalination have catalyzed a general revival of interest in membrane separation processes. It is not surprising, therefore, that attempts are currently underway to reinvestigate those aspects of membrane separations which have not been definitively answered in the past. One key problem area has been a dearth of direct morphological evidence concerning the relationship between structure and function in semipermeable membranes. Neither optical nor electron microscopy has been very successful in elucidating the structure of semipermeable membranes; the former because of low resolution, and the latter because of the tedious sample preparation, the introduction of artifacts, and the interpretation of extremely flat images. Owing partially to these difficulties the functional approach has tended to predominate in the councils of membrane technologists. Very recently, however, a new investigative tool, scanning electron microscopy [1], has become available which promises to add substance to the field of membrane structure. The scanning electron microscope exhibits resolution midway between optical and standard electron microscopes and offers great advantages over both in ease of sample preparation and in image depth and quality. These advantages are all intrinsic to the nature of the scanning electron microscope, which utilizes back-scattered and secondary electrons over a wide range of angles to obtain a reflection image which is scanned and transmitted to an oscilloscope for photographing. In this paper, the application of scanning electron microscopy to such representative membrane problems as swelling, asymmetry, void size, pore size, rugosity, and the morphological changes attendant upon hydrolysis and de-swelling will be considered together with some speculation as to its eventual application to liquid and nascent membranes.