Abstract It is generally acknowledged that the first waterflood occurred as a result of accidental water injection in the Pithole City area of Pennsylvania in 1865 (1). As long ago as 1880 it was concluded that water, finding its way into a wellbore from shallow sands, would move through oil sands and be beneficial in increasing oil recovery! Since that time waterflooding has become the dominant technique employed in world wide oil recovery operations. An area of particular interest at the present time is the use of sea water as an injection fluid, particularly in the North Sea, where the number of offshore oil production facilities is rapidly increasing. The high expense of offshore operations plus the high capital cost of the platform very often impose severe constraints on the design engineer, particularly in terms of the size and the weight of the water treatment facilities. The final design of any system will be governed by a variety of factors, generally with economics predominating. This paper discusses the influence of external constraints and reservoir criteria on offshore water injection system design. Examples are drawn mainly from BP's 5 years operational experience with the Forties Field Sea Water Injection System. The design of a waterflood project requires two major facets of work, the reservoir engineering aspects, to predict recovery performance by waterflooding, and the production engineering, to evaluate all the surface requirements. Work on the former is normally limited to reservoir simulation studies to determine the water injection volume and pattern requirements. An important factor in the success of a waterflood project, although often neglected in production engineering, is the quality of the water required for injection into the reservoir rock. This definition has three major considerations: