A miniature solid state dosimeter based on the phenomenon of thermoluminescence (1) is presently being investigated in our laboratory for introduction into an in vivo dosimetry system, previously described by the authors (1, 3, 4). The radiation-sensitive element is a manganese-activated calcium fluoride phosphor. Development of this phosphor was carried on at the U. S. Naval Research Laboratory by Schulman et al. (2). The dosimeter we are using is in the form of a needle, 10 mm. long and less than 1 mm. in outside diameter. It is sealed within a glass tube and read by heating at a constant rate. Heating liberates electrons which were trapped during irradiation, and these electrons attempt to return to their ground state by emitting energy in the form of blue-green light. A graph of fluorescent intensity vs. time, generated at a constant heating rate, is called a “glow curve.” Either the area under the glow curve, or its peak, may be used to read the dosimeters; we prefer to use the former. This area may be measured by electronically integrating the current output while the glow curve is traced. The most difficult problem in developing the reader system has proved to be the mounting of the photomultiplier assembly in the area of the heating circuit. The dosimeter is a relative device. It is calibrated by exposing a large number of dosimeters to different radiation modalities. Since the response is linear, the area under the glow curve increases as the exposure increases. A calibration factor having the dimensions of roentgens per unit area may thus be determined. The act of reading destroys the electron traps. Hence, this dosimeter cannot provide a permanent exposure record nor can it be used as a cumulative device. This fact plus its sensitivity and range distinguish it from the radiophotoluminescent dosimeter (3, 4). The range of the thermoluminescent dosimeter is estimated to be from 10−5 to 105 r, a span of ten decades. Like the radiophotoluminescent dosimeter (3), it is energy dependent and corrective shielding is required. The shielded and unshielded dosimeters will be subjected to x-rays from 260 kv to 1 Mv, cobalt 60, radium, and iridium 192 in air, and at various positions in depth of phantoms with varying portal sizes to determine the effect of energy on response. When satisfactory shields have been evolved, exposures will be made in animals. Thermoluminescent dosimeters developed to date have been of the large bulb type. Our work deals solely with miniature dosimeters similar to those being used in Discoverer satellites. Our objective is the development of a second generation of implant dosimeters for radiation dosage verification (4) and inhomogeneity studies with radiation therapy equipment as well as with diagnostic x-rays in specific anatomic areas, while patients undergo various examinations. The thermoluminescent dosimeter is intended to complement, not to supplant, the radiophotoluminescent dosimeter.