[Illustration: A. A. Line of Fluitation.] On several ships, some of the protectors, in the stem and the stern, were placed _vertically_; in which case they were fastened to the stems and stern-posts; and in this manner they were found to act more powerfully in preserving the copper, than when they were all placed horizontally. The ends of the protectors were rounded, in order to prevent any great resistance to the water, and they were fastened to the bottoms of the ships with copper bolts, the iron being counter-sunk to receive their heads, and the holes were then filled with carbonate of lime, or Parker's cement. To bring about the best possible contact of all the copper sheets, their edges, which lap over each other, where the nails are driven to fasten them to the ships, were rubbed bright, first with sand-paper, and finally with glass-paper. Shortly after the ships thus protected were sent to sea, it was evident to all on board, from their dull sailing, that the bottoms had become very foul; and on being examined in dry docks, it was found that the copper was completely covered with sea-weed, shell-fish of various kinds, and myriads of small marine insects. Upon their removal, however, it was found, on weighing the sheets, that the copper had suffered little or no loss; thus proving that, although its practical application had failed from unforeseen circumstances, the principle of protection was true, and had fully justified the expectation of its success. The copper near the protectors was much more foul than that at a greater distance from them; and there was, moreover, a considerable deposit of carbonate of lime, and of carbonate and hydrate of magnesia, in their vicinity. Sir Humphry Davy immediately suggested, as a remedy for this evil, that the bottoms should be scraped, and the copper washed with a small quantity of acidulous water; and he also proposed that the protectors should in future be placed under, instead of over, the copper sheathing. This plan was immediately adopted. Discs of cast-iron three and a half inches in diameter, and one-fourth of an inch in thickness, were let into the plank of the bottom of the Glasgow, of fifty guns, on the starboard side only--the larboard side having been left without any protection. These discs were in the proportion of one to every four sheets of copper, and over them were placed pieces of brown paper, and over the paper thin sheet-lead, so that the latter metal was in contact with the copper sheathing. A similar experiment was also tried on the Zebra, of eighteen guns, substituting, however, discs of zinc[106] for those of iron.