Following a review of methods for measuring the optical properties of tissue, the majority of this thesis is concerned with the design, construction, calibration and use of a solid, tissue equivalent phantom. The phantom material is a clear polyester plastic. This is obtained in unpolymerised form, scattering particles and absorbing dyes are added to it, and it is then polymerised to form a stable solid. Purely scattering and absorbing phantoms were made separately, and their optical properties were measured using a specially built system. This has a co-linear collimated light source and detector, and measures the unscattered light transmitted through a sample as a function of its thickness. Other methods of measuring the optical coefficients of tissue were tested with this phantom. One of these uses integrating spheres to measure the transmitted and reflected light from a sample. A model of light transport (in this case a Monte Carlo model) is used to convert these measurements into scattering and absorption coefficients. It was found that the measurement of scattering coefficient was reasonably accurate, but that the absorption coefficient was overestimated at the low values typical of tissue. A measurement of the optical properties of bone was made with this system. The other system investigated uses the diffusion theory to calculate optical properties from measurements made through a thick slab. The material was also employed to create a test phantom for near infrared spectroscopy machines. This provides a diffusing medium with an attenuation that is variable in discrete steps over three orders of magnitude. The relative attenuation between steps is totally wavelength independent. This phantom was adopted by the EC concerted action on near infrared spectroscopy and imaging. Finally, the phantom was used to create test objects with which to investigate the potential of imaging with infrared light.