The kinking of deep-sea cable during repairs, and the failure of the cable to coil properly when an endeavour is made to recover it, are troubles which are as old as the art. The paper attributes these troubles to two causes—the tendency of laid-up cable to unlay under tension, and the twisting action of the bowsheave when the cable trends outboardto port or star board. An example is given of a method of calculating the tendency to unlay under tension. It is shown what are the effects resulting from a trend of the cable to port, tostarboard, and when central; also that the variation of the effects when the trend is altered follows an exponential law. It is further shown that a certain trend of the cable outboard enables torsion in the suspended cable to be avoided during pauses in picking up, thus avoiding kinking at the bottom, and that the constancy of the trend conduces to satisfactory coiling. Novel forms of sheaves are described which would facilitate keeping the cable outboard at a constant trend relative to the sheave.