[Illustration:  FIG. 4_a_.—­GOVERNMENT SYSTEM OF LIGHTNING CONDUCTORS FOR LARGE BUILDINGS.]

[Illustration:  FIG. 4_b_.—­GOVERNMENT SYSTEM OF LIGHTNING CONDUCTORS FOR SMALL BUILDINGS.]

In Fig. 4 (a and b) is shown the Government method for protecting buildings in which explosives are made or stored.  Multiple points or aigrettes would be better.  Lord Kelvin and Professor Melsens favour points, and it is generally admitted that lightning does not strike buildings at a single point, but rather in a sheet; hence, in such cases, or in the event of the globular form being assumed by the lightning, the aigrette will constitute a much more effective protection than a single point.  As to the spacing of conductors, they may, even on the most important buildings, be spaced at intervals of 50 feet.  There will then be no point on the building more than 25 feet from the conductor.  This “25-feet rule” can be adhered to with advantage in all overground buildings for explosives.

Underground magazines should, whenever possible, also be protected, because, although less exposed than overground buildings, they frequently contain explosives packed in metal cases, and hence would present a line of smaller electrical resistance than the surrounding earth would offer to the lightning.  The conductor should be arranged on the same system as for overground buildings, but be applied to the surface of the ground over the magazines.

In all situations where several conductors are joined in one system, the vertical conductors should be connected both at the top and near the ground line.  The angles and the prominent portions of a building being the most liable to be struck, the conductors should be carried over and along these projections, and therefore along the ridges of the roof.  The conductors should be connected to any outside metal on the roofs and walls, and specially to the foot of rain-water pipes.

All the lightning conductors should be periodically tested, to see that they are in working condition, at least every three months, according to Mr Richard Anderson.  The object of the test is to determine the resistance of the earth-connection, and to localise any defective joints or parts in the conductors.  The best system of testing the conductors is to balance the resistance of each of the earths against the remainder of the system, from which the state of the earths may be inferred with sufficient accuracy for all practical purposes.

Captain Bucknill, R.E., has designed an instrument to test resistance which is based on the Post Office pattern resistance coil, and is capable of testing to approximate accuracy up to 200 ohms, and to measure roughly up to 2,000 ohms.  Mr R. Anderson’s apparatus is also very handy, consisting of a case containing three Leclanche cells, and a galvanometer with a “tangent” scale and certain standard resistances.  Some useful articles on the protection of buildings from lightning will be found in Arms and Explosives, July, August, and September 1892, and by Mr Anderson, Brit.  Assoc., 1878-80.