This thesis represents the investigation and development of neutron imaging techniques as a versatile method that can be used to reveal the presence, position, size, morphology, and chemical composition of structures and materials within a sample. The underlying physics of neutron radiography and tomography mean that it has fundamental similarities to x-ray techniques, although some subtle differences mean
that neutrons are often applicable in circumstances where x-rays are not. This complementarity is presented and discussed.
Most of the experimental results contained herein were obtained using Neutrograph, the thermal neutron imaging instrument at the Institut Laue-Langevin's high flux research nuclear reactor in Grenoble, France. Neutrograph utilises the world's most intense neutron beam currently in use for this purpose, -3xl09 n·cm-2·s-1 across an area of
220x220 mm2, and is capable of producing high quality images with single-frame exposure times down to the millisecond regime. Meanwhile, the modest beam
divergence of 6 mrad (equivalent to a beam length-to-diameter ratio, LID, of -150) can achieve a spatial resolution in the region of 150-300 11m.
It is demonstrated how these properties make Neutrograph ideal for obtaining timeresolved measurements of dynamic processes in both two and three dimensions, and for imaging relatively thick samples or highly attenuating materials on a reasonable time-scale. One aspect that is addressed in this research is the potential scope of relevance of neutron imaging methods across a range of scientific disciplines. In this vein, attempts have been made not only to draw examples from a number of different fields (palaeontology, archaeology, and fluid mechanics, for example), but also to improve the performance of the instrumentation and methodology, and to establish
novel techniques that will further the applicability of neutron imaging. An illustration of the latter is the use of a polarised neutron beam to investigate the presence and
distribution of magnetic components in non-magnetic matrices.