scattered into thousands of fragments, without any damage being done to the vessel itself. A Turkish ship cruising near at the time, with a chain from the masthead which did not reach into the sea, had a hole like that which would have been made by a cannon-shot pierced through the hull near the water-line. The inference was drawn from these cases that chains, and especially small chains, were not trustworthy for the purpose of conducting discharges of lightning. The mechanical violence sustained was perceived to be due to the circumstance that the conductors provided were of a bad principle of construction. They were at the least from nine to ten times too small. Conductors provided by engineering art are intended to be struck, but struck in such a manner as to govern the lightning, and to render the heaviest strokes harmless. No case had been known of a continuous iron rod, three quarters of an inch in diameter, or with a sectional area of one and a quarter square inch, having been structurally injured. The cases alluded to were held to demonstrate that conductors must have a sufficient size and thickness of metal, and must be continuous and without defect from end to end. It was definitely settled that, in accordance with these requirements, a square iron rod used as a defense against lightning should have, at least, a diameter of nine sixteenths of an inch, and that a round rod should have a diameter of ten sixteenths of an inch.
Some modification was also made in this instruction in reference to air-terminals. It was considered that a blunt point, fashioned like the apex of a cone subtending an angle of thirty degrees, would be less liable to fusion than a sharper and more attenuated point, and that therefore it should be adopted for the upper terminal, although it might, perhaps, not exert altogether so satisfactory a neutralizing influence. The area protected by a conductor was now considered not to be so definite and certain as it was previously held to be. It was recognized that it would be less in the case of a building with a metal roof, for instance, than in other circumstances. The earth contact, it was remarked, could not be looked upon as efficacious unless it were made, through the instrumentality of sheets of water, at least as large as the area of the storm-cloud, and access to such sheets must be secured by boring both in the direction of the surface moisture and in that of the deeper soil. Chains of red copper with a square section of three eighths of an inch, and weighing a pound and three quarters per yard, were recommended for ships. Such were the principal suggestions of a practical kind that were submitted in this report. In all other particulars the provisions of the earlier instructions were substantially approved and confirmed. There was, however, one incidental remark contained in this excellent report which is deserving of the highest commendation and approval on account of its practical wisdom. This emphasized the necessity for continued and minute observation and study of the effects of thunder-storms, with a view
