Once more the ‘Académie’ nominated a committee on lightning conductors, composed of the same members who had signed the ‘Instruction’ of 1854, and drawn up by Professor Pouillet. He again drew up the report, which was adopted by the ‘Académie’ on February 19, 1855, and contained some notable additions to the directions previously given. They related, as was desired, in the first instance to the Louvre alone, but were made applicable to all large public buildings. For their efficient protection, the professor insisted, two things should be kept in view above all others--namely, first, that the point, always of copper, should be of greater thickness; and, secondly, that it should have a never-failing connection with either water or very moist earth. To ensure the latter, it was recommended, as had been done before, that the underground part of the conductor should be divided. The necessity for such a division, and for forming at least two subterranean arms--the first of it, described as ‘the principal branch,’ going very deep into ground, into perennial water, and the second, ‘the secondary branch,’ running nearer the surface--was explained by Professor Pouillet very clearly in this last report. ‘After a long continuance of dry weather,’ he observed, ‘it often happens that the lightning-bearing clouds exert their influence only in a very feeble manner on a dry soil, which is a bad conductor; the whole energy of their action is reserved for the mass of water which by percolation has formed below it. It is here that the dispersion of the electric force (_la décomposition électrique_) takes place; it will follow the principal branch of the conductor underground, and leave the secondary branch untouched. The case is entirely different when, instead of dry weather, there have been heavy rains, moistening the earth thoroughly, up to the surface. It is the latter now that is the best, because the nearest, conductor of the electric force, which will not go to the more permanent sheet of water, lying more or less deep in the ground, if there is moisture above it. Under these circumstances, it is indispensable that there should be a direct connection between the surface soil and the lightning conductor, and this is what is accomplished by the secondary branch. It is a power in aid of the principal branch, and one often of the highest importance.’ The suggestion here made was one so evidently good, that it was at once accepted by the French Government, and the Louvre not only, but other public buildings, received lightning conductors ending in two subterranean branches, as proposed by Professor Pouillet.