While it is unusual to test tank-models with controlled, variable elasticity there are several reasons why we believe that it is im-portant to exert the large effort required. 1. The convenient model-building materials are much too stiff to represent the proper behaviour of concrete and steel at full scale. 2. We believe that, although the close-packed, crest-spanning ter-minator configuration is the best possible arrangement for a wave-energy device, it is not economic to resist bending moments great-er than those which would occur at the power limit. This means that non-destructive yielding to bigger waves must be achieved. This will not only save money on the spine structure but will also produce dramatic reductions in mooring forces. 3. Work in the narrow tank had shown that the correct control of a duck mounting could produce large improvements in duck perfor-mance, doubling the efficiency in waves twenty-five duck diameters in length. We had also discovered that the best values of mounting stiffness were low, less than those that would be provided by post-tensioned concrete at full scale, and that the mounting move-ment could itself be a useful power-generating mechanism. Further-more the hardware needed to provide non-destructive yielding needed very little modification to provide intelligent control and the extra generating capability. We realised that the software require-ments would be formidable - well beyond our present knowledge. But we were confident that the existence of a controlled model would, as so often before, stimulate theoretical work. Furthermore we were quite certain that the progress of computer technology be-tween now and the date of the return of energy shortage would be so enormous that any level of control sophistication could be safe-ly assumed to be available at virtually zero cost. On many occa-sions it has seemed that this view was not shared by the civil and marine engineers who assess our progress. The work reported in this volume covers the measurements of bending moments and joint angles as a function of sea conditions, model lengths and stiffness, for circular spine sections without ducks. Other volumes will contain the observations of shear, axial, torsion and mooring forces, for spines with various appendages.