The Earth Resources Program relies on the fundamental premise that remote sensing can provide knowledge of ground conditions over large areas, and possibly all, of the earth's surface. The collection of this information, by means of radiant energy sensors in aircraft or satellites, depends on two important factors:(1) that all objects on the earth's surface continuously radiate or reflect electromagnetic energy; and (2) that spectrally-selective emission and reflection can uniquely identify objects of interest, with "spectral signatures" serving as the leading means of identification. A major facet of the problem of translating this basic concept to practice is that of data correlation. In the most general case, large quantities of data are collected from a variety of sensors, over a broad range of wavelengths, and from different gyrating vehicle platforms at different altitudes and times. Automatic correlation of the data from one collection to another, with other reference data, and with the ground is necessary if the acquired information is to be useful in providing a basis for subsequent corrective action. This paper discusses a pilot system design for an Automatic Data Correlation System which was the subject of a year long study performed for NASA. The Pilot version, although intended to be primarily a research and development tool, provides all the essential functional capabilities of the ultimate, advanced system. It was designed as an inexpensive, easily assembled, configuration of required equipments that allows thorough experimentation with all stages of the needed film and magnetic tape data handling while accomodating all presently identified Earth Resources remote sensors.