During a rocket flight to high altitude (585 km) a narrow band signal on an electric antenna was sometimes observed, whose frequency varied as the rocket turned. Such a signal cannot be natural, but apparently must be generated by the interaction of the rocket-antenna system with the ambient plasma. Conditions for development of the instability are investigated. Maximum oscillation amplitude occurs when the antenna is aligned with the Earth's magnetic field. Similar observations have been reported by Gurnett and Mosier [1969>. Several attempts are made to understand the nature of this interaction, but without success. The instability treated by Fiala, due to the interaction of an inductive antenna impedance with stray capacitance to a phase shifted point in the preamplifier, can be ruled out. It appears that a negative antenna resistance due to interaction with waves Doppler-shifted through zero is an unlikely explanation. The rocket velocity seems too small to give such an anomalous Doppler shift, and even if the observations of plasma density and temperature are stretched, the positive sheath resistance is larger than calculated negative resistances. Ion transit time instability in the sheath would only work at 10 times higher frequency. Interaction of the flowing plasma with sheath waves around the antenna is suggested and appears promising but the theory is not sufficiently developed for meaningful comparison. ¿ American Geophysical Union 1990