This book is at once the epic and the encyclopaedia of whaling. It is a monument to the honour of an extinct race of daring seamen; but it is a monument overgrown with the lichen of neglect. Those who will care to scrape away the moss may be few, but they will have their reward. To the class of gentlemanadventurer, to those who love both books and free life under the wide and open sky, it must always appeal. Melville takes rank with Borrow, and Jefferies, and Thoreau, and Sir Richard Burton; and his place in this brotherhood of notables is not the lowest. Those who feel the salt in their blood that draws them time and again out of the city to the wharves and the ships, almost without their knowledge or their will; those who feel the irresistible lure of the spring, away from the cramped and noisy town, up the long road to the peaceful companionship of the awaking earth and the untainted sky; all those-and they are many-will find in Melville's great book an ever fresh and constant charm.

Dalhousie College, Halifax, N.S.

ARCHIBALD MACMECHAN.

LIGHTNING RODS.

LIGHTNING being known to be a manifestation of electricity,

the protection of objects from its effects became a problem in the application of the laws of electricity. The history of lightning rods might therefore have been expected to resemble, or at least run parallel with the history of other branches of electricity; but such has not been the case. The fact is that Franklin, two years before his famous kite experiment had demonstrated the nature of lightning, devised a method of protection which has been pronounced perfect by experts within the last few years. Franklin's experiment and his interpretation of it deserve description in his own language* :

"Take a pair of large brass scales, of two or more feet beam, the cords of the scales being silk. Suspend the beam by a thread from the ceiling, so the bottom of the scales may be about a foot from Condensed from an account sent to Peter Collinson, July 29, 1750.

the floor; the scales will move round in a circle by the untwisting of the thread. Set an iron punch on the end upon the floor, in such a place as that the scales may pass over it in making their circle; then electrify one scale. As they move round, you see that scale draw nigher to the floor, and dip more when it comes over the punch; and if that be placed at a proper distance, the scale will snap and discharge its fire into it. But if a needle be placed upon the fluor near the punch, its point upwards, the scale, instead of drawing nigh to the punch and snapping, discharges its fire silently through the point, and rises higher from the punch.

Now if the fire of electricity and that of lightning be the same, these scales may represent electrified clouds. The horizontal motion of the scales over the floor may represent the motion of the clonds over the earth; and the erect iron punch, a hill or high building; and then we see how electrified clouds passing over hills or high buildings at too great a height to strike may be attracted lower till within their striking distance. And lastly, if a needle fixed on the punch with its point upright, or even on the floor below the punch will draw the fire from the scale silently at a much greater than the striking distance, and the punch is thereby secured from the stroke; may not the knowledge of this power of points be of use to mankind, in preserving houses, churches, ships, etc., from the stroke of lightning, by directing us to fix on the highest parts of those edifices upright rods of iron, made sharp as a needle, and gilt to prevent rusting, and from the foot of those rods a wire down the outside of the building into the ground."

Over the details of this protective system controversies have been numerous. Should the conductor be iron or copper, wire, rod, ribbon or cable? Should it terminate in a point or a ball? (In England points, because favoured by Franklin, were supposed to "savour of republicanism.") How high should the rod rise above the roof of the building it protects and over what area does its protection extend? Some of these questions are now regarded as unimportant, others are still under discussion, while others again have been completely answered by experience. The answers may be found in the reports of various committees of the French Academy of Sciences, Lightning Rod Conferences and other bodies where minute directions are given regarding the size and construction of each part of the conductor. But regarding the principle of Franklin's method of protection there has been, until very recently, but one opinion-that it is perfect.

"The testimony of electrical engineers who have had large experience with lightning conductors seems almost unanimous that a lightning conductor erected and maintained in accordance with the conditions prescribed by the Lightning Rod Conference gives perfect protection."

This should surely be sufficient answer to the question so often asked: Are lightning rods of any use? For the question is obviously one which cannot be answered by a reference to a few cases. The opinion of men of long experience is that the value of conductors in preventing damage from lightning can scarcely be over-estimated, and this opinion is endorsed by scientists who have made a systematic study of thousands of recorded facts.

A lightning rod protects the object upon which it is erected in two ways. The point by promoting silent discharge relieves the cloud of its charge without the destructive effects of a flash. The quantity of electricity thus drawn from clouds by a point may be very considerable. Beccaria by breaking a conductor leading from one of the seven roofs of the Valentino Palace at Turin and watching the sparks cross the gap estimated that during one thunderstorm that conductor drew from the clouds in an hour enough electricity to kill 360 men. This is of course very indefinite, but it means that he obtained across the spark-gap what to the eye and ear seemed a continuous series of powerful sparks, each of which would have given a painful shock. It cannot be doubted that during that hour the conductors on those seven roofs drew from the clouds enough electricity to prevent many destructive flashes. Numerous observations of the behav iour of thunderclouds before and after passing over villages or castles well equipped with pointed lightning rods strengthen this opinion.

In the second place, if a lightning flash does occur, it is expected to strike the point of the conductor and pass quietly to earth through the metal. This it generally does, so generally indeed that at the meeting of the British Association in 1888, so eminent an authority as Mr. Preece endorsed the statement of the Lightning Rod Conference of 1882 that "there is no authentic case on record where a properly constructed conductor failed to do its work." This statement however is too strong, as carefully constructed conductors sometimes fail. When they do, the

dogmatic assertion that there must have been some undiscovered flaw in their construction is thoroughly unscientific. Unfortunately such assertions have frequently been made. The truism that "lightning always follows the path of least resistance," has been taken to mean that "lightning always follows the path whose resistance to ordinary electric currents is least." The resistance, in this sense, of a stout metal rod running down from the top of a building and making good connection with the earth is necessarily less than that of any other path from the top of that building to the ground. Therefore, if lightning strikes the conductor of that building, and, after following it some distance, leaves it to jump through a brick wall and run along some bell wires, or do other damage, it is held that there must have been some fault in the construction of the rod or in its connection with the earth. The earth-connection being out of sight is generally blamed. Now this conception of the action of a conductor is wrong, and this method of accounting for damage to protected buildings has done much to retard scientific knowledge of lightning by fixing attention upon imaginary faults in the rod, or dryness of the ground when the phenomenon was really due to an unsuspected property of lightning. Thus it was not until Professor Oliver Lodge showed by experiment* the effects of selfinduction upon a discharge passing along a wire in the laboratory that attention was directed to any other property of a lightning conductor than its resistance.

The history of the Washington monument furnishes a good illustration of the occasional failure of lightning to strike the highest point of a conductor. The apex of the pyramidal top of this monument is a block of aluminium, which is connected by large wires to the steel columns of the elevator shaft, which are in turn well connected with the earth. The whole forms an ideal lightning rod 555 feet high. Most of the discharges which occurred near the monument no doubt struck the aluminium point, and were carried away unnoticed, but one struck the monument several feet from the point and made its way through the stonework to the elevator shaft, doing much damage.

The following quotationt illustrates the failure of lightning to follow a conductor even after it has reached it':

*See his Lightning Conductors and Lightning Guards to which this article is indebted at many points.

+Gerald Molloy Lightning, Thunder and Lightning Conductors.

"In the month of May, 1879, the Church of Laughton-en-le Morthen, in England, though provided with a conductor, was struck by lightning and sustained considerable damage. On examination it was found that the lightning followed the conductor down along the spire as far as the roof; then changing its course, it forced its way through a buttress of massive mason work, dislodging about two cartloads of stones, and leaped over to the leads of the roof, about six feet distant. It now followed the leads until it came to the castiron down-pipes intended to discharge the rain-water, and through these it descended to the earth."

These are two out of very many such cases, all of which have been held to be explained by saying that there was some defect in the earth connection. But any explanation based upon an assumed defect in the earth connection is insufficient, for in both the cases mentioned and in most of the others, even if the conductor merely touched the surface of the ground, it formed a path of smaller electrical resistance than that chosen by the lightning. It must then be admitted that even the best constructed lightning rods occasionally fail. The lightning may not strike the point, or a part of the discharge may leave the conductor after following it for some distance. Both of these phenomena are capable of scientific explanation.

The effect of a rod cannot extend to a distance many times its own dimensions. It is idle therefore to suppose that in determining the path of a flash a mile long, a rod of a few yards. length can exercise a controlling influence. In comparison with the magnitude of the flash the building or tower bears a closer resemblance to the needle-point which receives a spark in the laboratory than the pointed conductor which it carries. More than that, the same causes which so frequently divide a flash in the air must operate in the same way to prevent the whole of a heavy flash from striking the slender point of a conductor. The idea of an "area of protection," i.e. of a certain region round a rod within which it is impossible for lightning to strike, is therefore absolutely wrong. It is, in fact, quite possible for lightning to strike the top and bottom of a rod at the same time.

The object to be sought in protecting a large building is not to erect so large a rod that the lightning will be sure to strike it, but to place conductors over all prominent parts so that wherever