This study covers the design, construction and testing of a wind propelled hydrofoil trimaran. The work was split into four main parts which when integrated as a whole formed the basis of the information required for a design to be formulated. Much of the work and the computer programs which were written are applicable to hydrofoil craft in general (both motor powered and wind propelled) and because of the type of hydrofoil design considered, they are mainly orientated towards surface piercing hydrofoil systems.

A prototype boat of 5 metres overall length was designed and built in order that full scale-tests could be carried out on the open water. Most of the results from these tests were either qualitative or photographic. Even so the results from these trials were a most useful indication of the performance of the design. Structural problems were encountered with the construction of the hydrofoils.

Computer programs were written to predict the calm water steady state lift and drag, and flight orientation of the hydrofoil boat. These calculations included predictions at angles of heel and yaw. The results from these 'predictions were compared with a series of model tests undertaken on a one quarter scale model of the 5 metre prototype. Agreement was found to be good.

Analysis methods were formulated and predictions obtained for two different wind propulsion systems, a soft sail rig and a horizontal axis wind turbine rig. The soft sail rig was used on the prototype boat, but the turbine was shown to offer scope for a more versatile propulsion system if exceptionally high speeds were not aimed for. High boat speeds in low wind speeds, unfortunately only over a limited range of courses relative to the wind direction, were best obtained by resort to a soft sail or solid aerofoil rig. Consideration was given to the operation of the wind turbine both in the windmill mode and the autogyro mode.

A theoretical study was made of the seakeeping of surface piercing hydrofoil systems in regular head and following seas. This incorporated a linearised solution in the frequency domain and a non-linear step-by-step simulation in the time domain which was executed on a digital computer. From a comparison of the results, of these'simulations with a series of model experiments, it was found that the latter method gave the best solution in head seas and offered the best possibilities for an accurate solution in following seas.