This thesis relates to the condition monitoring of long steel pipes using acoustic emission (AE). A number of experiments were carried out on pipes with a range of internal and external environments using a linear axial array of sensors with the ultimate aim of locating and reconstituting the time-domain and frequency-domain signatures of AE sources. The AE waves were generated from simulated, discontinuous, continuous, and semi-continuous sources and from real sources generated by impacts and crack propagation. The simulated source work in different internal and external environments was carried out to develop a generic empirical approach to AE propagation in long steel pipes which acknowledges the distortion of a source disturbance in the time and frequency domains. Generally, the acquired signals have two identifiable components and methods are developed for separating these components automatically and determining their group velocities. A simple model for attenuation is also developed which includes effects brought about by burial of the pipe and /or the nature of the fluid transported (liquid or gas). In the impact and crack extension tests a variety of intensities were simulated and the effect of type and intensity on time- and frequency-domain characteristics of the source was determined. The overall outcome is the demonstration of the potential of AE for identifying the nature, intensity and location of damaging events, such as crack growth and denting, and for the location and intensity assessment of leaks.