It is with due humility that we venture to discuss supervoltage roentgen therapy within the shadows of San Francisco and Berkeley. With the advent of newer forms of high-voltage accelerators, which seem to place no finite limits upon the energy of x-ray or particle beams, rays generated within the range of 1,000 kv. can now scarcely be classed as supervoltage. We make no apology, however, for attempting to evaluate our ten years of experience with radiation of this modest energy (1). In so doing we are brought face to face with the ever more striking fact that our biologic and clinical knowledge lags behind. technical development by many orders of megavolts. It is now approximately twenty-six years since Coutard gave to roentgen therapy the status of an independent and useful art. His medium-voltage equipment, archaic by present standards, provided him with a tool which enabled him to make radiobiologic and clinical observations, obtaining results still rarely surpassed and infrequently equalled by those using modern apparatus. It cannot be said that we as yet fully understand biologic effects and clinical application of x-rays within the range of 200 kv. X-rays within the range of 1,000 kv. have had clinical application for about fifteen years. Our own personal experience covers ten years. We have had the advantage in our own clinic of drawing upon the patients and records of our predecessor, Dr. John E. Wirth, who encountered some of the initial pitfalls in the use of supervoltage x-rays. Controlled clinical appraisal in the Tumor Institute of the Swedish Hospital, therefore, dates from 1935, covering a thirteen-year period. Apparatus and Technical Factors The apparatus which we have used consists of a General Electric KXC-2 installation, which was engineered in 1932, the pioneer period in the development of x-ray voltages of about 500 kv. The cascade banks of transformers and condensers and the operation of the tube (14 feet in length) in air necessitate a large shielded room to provide clearance between the voltage-generating apparatus. The system is now archaic in design when compared with modern apparatus; yet its performance has been remarkable. The time loss due to repairs and maintenance has not appreciably exceeded that of more conventional medium-voltage equipment. Operational voltage has been maintained at 800 kv. as recorded by rotary voltmeter measurements (2). Provision for controllable cathode bias (3) with the filament operating at 10 rna. produces a beam of effectively higher quality for a given applied potential. Filtration of 4.25 mm. lead in addition to secondary filters gives a beam of half-value layer equivalent to 3 mm. lead or 9.1 mm. copper. The effective wave length has an average value of 0.0245 Å. Target-skin distance is 100 cm. Cones are avoided by the interposition of a light-centering and field-defining mechanism, the beam traversing the last 30 cm. in air before striking the skin.