Two experiments have been conducted to determine the accuracy with which an aircraft can be positioned in flight using carrier phase measurements from satellites of the NAVSTAR Global Positioning System (GPS). In each experiment, a photogrammetric cameras was flown over a calibration area providing ground control that enabled aerotriangulation to be performed on ground targets or, conversely, for the camera position to be independently determined. GPS measurements were used to determine the position of an antenna located on the aircraft at the time of each photo and, using the fixed antenna-camera vector, to determine the positions of the camera. The GPS-derived camera positions were then used to compute the positions of the ground targets. In an experiment in Texas the differences between the conventional and GPS-based aerotriangulation display biases of 1--3 cm horizontally and 28 cm vertically and RMS differences of about 2 cm, 3 cm, and 6 cm in north, east, and height, respectively. The camera positions derived from the aerotriangulation showed similar biases when compared to the GPS-derived camera positions and RMS differences of about 7 cm in the horizontal and 9 cm in height. Because of edge effects in the photography the actual precision is slightly better. A second experiment in Washington revealed biases of several meters, apparently due to unresolved coordinate system differences. However, a Helmert transformation in this case produced RMS differences between GPS-derived and photogrammetrically determined ground positions of 3--5 cm. While these results demonstrate the need for care in relating GPS results to other geodetic measurements, they also show the great potential of GPS carrier phase measurements for precise kinematic positioning and a variety of remote sensing applications. ¿ American Geophysical Union 1989