This thesis describes the introduction of whole body Nuclear Magnetic Resonance (NMR) Imaging into clinical practice using the 1.7 MHz NMR imager designed and built in the Department of Bio-Medical Physics and Bio-Engineering at the University of Aberdeen. At the time of its introduction in 1980, it was the only such imager in the world capable of examining the whole body. In 1983, a 3.4 MHz imager was built in the Aberdeen Royal Infirmary by the same Department and the clinical trial continued using this larger instrument. NMR imaging is unique as an imaging technique in that it does not use ionising radiation to process images, but instead utilises the response of hydrogen protons in a magnetic field to pulsed radio-frequency signals, providing information about the body's soft tissues based on their water content. Studies to assess the usefulness of this imaging method for clinical diagnosis have been performed, paying particular attention to the measurement of proton-spin-lattice relaxation time (T<SUB>1</SUB>) which is known to vary in different disease states. It is shown that whilst the specific measurement of T<SUB>1</SUB> is not an accurate method for disease diagnosis, no one T<SUB>1</SUB> value being pathognomic of one disease state, the use of images made from T<SUB>1</SUB> measurements provide diagnostically useful information at both 1.7 and 3.4 MHz. The method has been compared with all other available diagnostic techniques, including tissue histology and is found to be a superior method for the examination of the cerebellum, brainstem, cervical cord and the base of the skull. It is as diagnostically useful as other diagnostic imaging methods for the study of malignancy in the head and neck region, the pelvis and musculo-skeletal system and in certain instances for the examination of the thorax and abdomen. Being free of ionising radiation, it has been found useful in the assessment of normal and abnormal pregnancy. NMR imaging at both 1.7 and 3.4 MHz provides a new, non-invasive method for the display of normal living anatomy and for the accurate diagnosis of a wide range of diseases. It is concluded that it will become an important new diagnostic imaging method, replacing some of the established methods of diagnostic imaging in clinical practice.