The hull of a timber-built ship consisted of a number of rigidly jointed frames or cells, some lying in horizontal, some in vertical, and some in intermediate planes: the unit cell being a quadrilateral, whose sides were formed by the frames and vertical riders and by the planks, wales, and horizontal riders. Practically all the materials composing the fabric of a ship were disposed either in planes parallel to the plane of the keel or in planes at right angles to it. And up to the end of the Napoleonic wars our ships, without appreciable exception, were built on this primitive quadrilateral system. The system was essentially weak. All warships showed a tendency to arch or hog--to become convex upwards, in the direction of their length--owing to the fact that the support which they derived from the water was relatively greater amidships than in the neighbourhood of their extremities. In the old days when ships were short in length this tendency was small, or, if appreciable, a remedy was found in working into the structures additional longitudinal and transverse riders, until the holds were not infrequently clogged with timber. But as ships increased in length, the forces tending to “break the sheer” of a ship and arch its keel increased in greater ratio than the ship’s power of resistance to the distortion; and by the end of the eighteenth century, in spite of the aid of iron knees, stronger fastenings, and improved material generally, the essential weakness of our mode of construction had been gradually exposed. The _Victory_ herself suffered from arching. The extremities of a 74-gun ship dropped six inches, sometimes, when she entered the water from the stocks. A similar tendency to hog took place also across the breadth of a ship, occasioned by the dead weight of her guns. When rolling in heavy weather the momentum of her top weights caused large racking stresses to be thrown on the joints between the frames and the deck-beams. The biographer of Admiral Symonds quotes Captain Brenton as follows: “I remember very well, when I was a midshipman in a 64-gun ship coming home from India, cracking nuts by the working of the ship. We put them in under the knees, as she rolled one way, and snatched them out as she rolled back again.” [Illustration: DIAGRAM ILLUSTRATING DISTORTION OF FRAMES UNDER LOAD]