dicularly, it is a certain indication that the clouds are at that height from which they may readily strike into the ground. It is also supposed that the appearance of two flashes at once indicates danger; but this is uncertain, as it may arise from an optical illusion, or the interposition of a

cloud.

307. The precautions generally offered by writers on electricity, as necessary for the safety of those who are exposed to the action of a thunder-storm, may be thus enumerated. Shelter should not be taken under trees, hedges, &c.; for, should they be struck, such situations are particularly dangerous; at the same time a person is much safer at about thirty or forty feet from such objects than at a greater distance, as they are likely to operate as conductors. Large portions of water also ought to be carefully avoided if possible, and even streamlets that may have resulted from recent rain; these are good conductors, and the height of a human being connected with them, may sometimes determine the course of the lightning. In a house the safest situation is considered to be the middle of the room; and this situation may be rendered still more secure bystanding on a glass-legged stool: but, as such an article is not in the possession of many people, a hair mattress, or a thick woollen hearth rug makes a very good substitute. It is very injudicious to take refuge, as some persons do, in the cellar during a thunder storm, since the discharge is often found to be from the earth to the clouds, and many instances are recorded of buildings that were struck having sustained the greatest injury about the basement story. But, whatever situation is chosen, the greatest care should be taken to avoid going near the fire-place, since the chimneys are most likely to attract the fluid, and even if there be no fire in the grate, at the time, it should be remembered that soot is a powerful conductor. The same caution is necessary with respect to all large metallic surfaces; gilt furniture, bell wires, &c.

308. But the most important and useful application that has ever yet been made of the discoveries of the electrician is in the method of securing buildings, ships, &c., from the effects of lightning. To the ingenuity of Dr. Franklin the world is indebted for this invention, as well as for his discovery of the identity of lightning and common electricity. He first proposed to erect a perfectly continuous metallic rod by the side of any building intended to be secured from injury by lightning: this rod was to be pointed at each extremity; to rise some feet above the highest part of the building, and to extend some feet below the foundation, in moist ground, or water if practicable. By such a precaution, he conceived that the house could never receive any damage; for, whenever the lightning should happen to fall upon it, it is evident that the conductor, being of metal, and higher than any part of the building, would naturally attract it, and, by conducting it to the ground, prevent that building from receiving any damage; it being well known that electricity always strikes the nearest and best conductors.

309. As this subject is of the utmost importance, we shall here quote the very concise and

judicious observations of Mr. Cava o respecting it. Speaking of the proposal of Dr. Franklin, which we have just noticed, he proceeds to remark, that the reasonableness and truth of such an assertion has been confirmed by numberless facts, and the practice of raising such conductors has been found exceedingly useful, 'particularly in hot climates, where thunder storms are very frequent, and the damage occasioned by the same too often experienced.

310. In regard to the construction of such conductors, there have been some controversies among electricians; and the most advantageous manner of using them has not, without a great many experiments, and but very lately, been ascertained. Some philosophers have asserted, that such conductors should terminate in an obtuse end, that they might the less invite the lightning from the clouds; for such an end will not attract electricity from so great a distance as a sharp point. But other philosophers have thought a pointed termination to be much preferable to an obtuse one; and their assertion seems, on the following accounts, to be better founded.

6

311. A sharp-pointed conductor, it is true, will attract electricity from a greater distance than one with an obtuse point, but at the same time will attract and conduct it very gradually, or rather by a continued stream, in which manner a remarkably small conductor is capable of conducting a very great quantity of electricity; whereas an obtusely terminated conductor attracts the electricity in a full separate body, or explosion, by which it is often made red-hot, melted, and even exploded in smoke, and by such a quantity of electricity as perhaps would not have at all affected it, if it had been sharply pointed.

312. A sharp-pointed conductor certainly invites the matter of lightning easier than an obtuse one; but to invite, receive, and conduct it in small quantities, never endangers the conductor; and the object of fixing a conductor to a house, is to protect the house from the effects of, and not the conductor from transmitting the lightning.

313. It is an observation much in favor of sharp-pointed conductors, that such steeples of churches, and edifices in general, as are terminated by pointed metallic ornaments, have very seldom been known to be struck by lightning whereas others that have flat terminations and have a great quantity of metal in a manner insulated on the tops, are often struck; and it is but seldom that they escape without great damage.

6

314. Besides these considerations a sharppointed conductor, by the same property of attracting electricity more readily than an obtusely terminated one, may prevent a stroke of lightning, which the latter is incapable of doing.

315. A conductor, therefore, to guard a building, as it is now commonly used, should, from several considerations and experiments, consist of one iron rod about three quarters of an inch thick, fastened to the wall of the building, not by iron cramps, but by pieces of wood. If this conductor were quite detached from the building, and supported by wooden posts at

the distance of one or two feet from the wall, it would be much better for common edifices, but it is more particularly advisable for powdermagazines, powder-mills, and all such buildings as contain combustibles ready to take fire. The upper end of the conductor should be terminated in a pyramidal form, with the edges, as well as the point, very sharp: and if the conductor be of iron, it should be gilt, or painted, for two or three feet. This sharp end should be elevated above the highest part of the building (as above a stack of chimneys, to which it may be fastened) at least five or six feet. The lower end should go five or six feet into the ground, and in a direction leading from the foundation; or it would be better to connect it with the nearest piece of water, if any be at hand. If this conductor, on account of the difficulty of adapting it to the form of the building, cannot conveniently be made of one rod, then care should be taken, that where the pieces meet, they be made to come in as perfect a contact with one another as possible; for, as we observed before, electricity finds considerable obstruction where the conductor is interrupted.

316. For an edifice of a moderate size, one conductor of the kind already described, is perhaps sufficient; but, in order to secure a large building from sustaining any damage by lightning, there should be two, three, or more conductors, in proportion to the extent of the building.

317. In ships a chain has often been used for this purpose, which, on account of its pliableness, has been found very convenient, and easy to be managed among the rigging of the vessel; but, as the electricity finds a great obstruction in going through the several links, so that chains have been actually broken by the lightning, their use is now almost entirely laid aside; and, in their stead, copper wires a little thicker than a goose-quill have been substituted, and found to answer extremely well. One of these wires should be elevated two or three feet above the highest mast in the vessel; this should be continued down the mast, as far as the deck, where, by bending, it should be adapted to the surface of such parts, over which it may most conveniently be placed, and by continuing it down the side of the vessel, it should be always made to communicate with the water.'

318. M. Cavallo gives, with cordial approbation, the following extract from the earl of Stanhope's learned work on electricity:-As requisites for the proper construction of conducting rods for the preservation of buildings from the effects of lightning he directs, (1.) That the rods be made of such substances as are the best conductors of electricity. (2.) That the rods be uninterrupted and perfectly continuous. (3.) That they be of a sufficient thickness. (4.) That they be perfectly connected with the common stock. (5.) That the upper extremity of the rods be as acutery pointed as possible. (6.) That it be very finely tapered. (7.) That it be prominent. (8.) That each rod be carried in the shortest convement direction, from the point at its upper end to the common stock. (9.) That there be neither large nor prominent bodies of metal upon the top of the building proposed to be secured, but

such as are connected with the conductor by some proper metallic communication. (10.) That there be a sufficient number of high and pointed rods: and, (11.) That every part of the building be very substantially erected.'

319. For the purpose of illustrating the principle on which the thunder-rod acts, there are some very satisfactory experiments generally employed by lecturers on electricity, which it may be proper here to explain, and before doing which we would just observe, that the charge used on these occasions should be moderate, having found from experience that, in that part of the experiments which is intended to show the effect of interruptions in the conductor, a strong charge has injured the apparatus.

320. The simplest form in which these experiments are made is that known by the name of the thunder-house. This is represented by fig 10, where A is a board about three-quarters of an inch thick, and shaped like the gable end of a house. It is fixed perpendicularly upon the bottom board B, upon which the perpendicular glass pillar C D is also fixed, in a hole about eight inches distant from the base of the board A. A square hole, I L M K, about a quarter of an inch deep, and nearly one inch wide, is made in the board A, and is filled with a square piece of wood, nearly of the same dimensions. This board must fit in rather easily, so that the slightest shaking may throw it out. A wire, L K, is fastened diagonally to this square piece of wood. Another wire, I H, of the same thickness, having a brass ball, H, screwed on its pointed extremity, is fastened upon the board A; so also is the wire MN, which is formed into a hook at (. From the upper extremity of the glass pillar, CD, a crooked wire proceeds, having a spring socket F, through which a double-knobbed wire slides perpendicularly, the lower knob G of which falls just above the knob H. The glass pillar DC must not be fixed very tightly into the bottom board; but it must be fixed so as to be pretty easily moved round its own axis, by which means the brass ball G may be brought nearer or farther from the ball H, without touching the part of EFG. When the square piece of wood LMIK, is fixed into the hole so that the wire LK stands in the dotted representation IM, then the metallic communication from H to O is continuous, and the instrument represents a house furnished with a proper metallic conductor; but if the square piece of wood LMIK is fixed so that the wire LK stands in the direction L K, as represented in the figure, then the metallic conductor HO, from the top of the house to its bottom, is interrupted at IM, in which case the house is not properly secured.

320.* Fix the piece of wood, L MIK, so that its wire may be as represented in the figure, in which case the metallic conductor HO is discontinued. Let the ball. G be fixed at about half an inch perpendicular distance from the ball H, then, by turning the glass pillar D C, remove the former ball from the latter; by means of a chain connect the wire E F with the wire Q of the jar P, and let another chain, fastened to the hook O, touch the outside coating of the jar. Connect the wire Q with the prime conductor,

and charge the jar; then, by turning the glass pillar DC, let the ball G come gradually near the ball H, and when they are arrived sufficiently near one another, the jar will explode, and the piece of wood, LMIK, will be driven out of the hole to a considerable distance from the thunder-house. In this experiment the ball G represents an electrified cloud; which, when it is arrived sufficiently near the top of the house A, the electricity strikes it; and, as this house is not secured with a proper conductor, the explosion breaks part of it, as is seen by the violent removal of the piece of wood I M.

321. Repeat the experiment with only this variation, viz. that the piece of wood IM is situated so that the wire L K may stand in the direction I M, in which case the conductor HO is not discontinued; and then the explosion will have no effect upon the piece of wood LM, this remaining in the hole unmoved; which shows the usefulness of the metallic conductor.

322. Again, unscrew the brass ball H from the wire HI, so that this may remain pointed. With this difference only in the apparatus, repeat both the above experiments, and it will be found that the piece of wood I M is in neither case moved from its place, nor will any explosion be heard; which not only demonstrates the preference of the conductors with pointed terminations to those with obtuse ends; but also shows that a house furnished with sharp terminations, although not furnished with a regular conductor, is almost sufficiently guarded against the effects of lightning.

323. This apparatus is sometimes made in the shape of a house, as represented by fig. 11; for the sake of distinctness, the side and part of the gable end AC represents that of the thunderhouse, and may be used in the same manner with that above described, or more readily by the following method:-Let one ball of the discharging rod touch the ball of the charged jar, and the other knob A of the conductor AC of the thunder-house; the jar will then explode, and the charge will act upon the conductor just mentioned. The conducting wire at the windows h, h, must be placed in a line. The sides and gable, A C, of the house, are connected with the lower part of the house by hinges, and the building is kept together by a ridge on the roof.

324. To use this model, fill the small tube a with gunpowder, and ram the wire c a little way into the tube; then connect the tube e with a large jar or battery. When the jar is charged, form a communication from the hook at C, on the outside, to the top of the jar, by the discharging rod; the discharge will fire the powder, and the explosion will throw off the roof, with the sides, back, and frout, so that they will all fall down together. Fig. 12 represents a small ramrod for the tube a, and fig. 13 a pricker for the touch-hole at C. Philosophical and mathematical instrument makers now construct the front of the common thunder-houses as well as the powder-houses above described, with two pieces of wood or windows, which, by being placed in proper situations, the one to conduct and the other to resist the fluid, will illustrate by one discharge the usefulness of good conductors for

securing buildings or magazines from the explosion of thunder, as well as the danger of using imperfect ones.

325. The most elegant method of performing this experiment, however, is represented in fig. 14, which exhibits a hollow pyrainid of wood, composed of several pieces, having a wire through each, so that their ends may come in contact with each other, as seen at s, s, s. One corner of the pyramid must be made loose as shown at d, having the conducting wire passing nearly through it, but not quite so. The wire, passing through the rest of the pyramid, must join (by a chain) the outside coating of a Leyden jar. If the cloud r be supported by a wire from the prime conductor, and hang half an inch from the knob g of the pyramid; when the jar is charged, a flash will take place between r and g; the fluid will pass along the wires s, s, s, till it comes to the break at d; there an explo sion will take place, that will force out the corner. piece d, and throw down the fabric in separate pieces.

326. The preceding experiments sufficiently illustrate the use of conductors raised on buildings that are much exposed to the effects of thunder-storms; but, that we may not omit any information on the subject that may be deemed useful, we shall just add Mr. Morgan's method of preventing all possible danger in these cases. The plan which Mr. Morgan proposes is that, whilst a house is being built, the foundation of each partition wall should be laid on a strip of lead, or that the strip be fastened to the sides of these partition walls. The strips should be two inches wide, and at least a quarter of an inch thick, and closely connected with each other. A perpendicular strip, on each side of the house, should rise from the conductors to the surface of the ground; whence a strip should be continued round the house, and carefully connected with water-pipes, &c. The strips on the sides of the house should then be continued to the roof, which ought to be guarded in the same manner as the foundation. The top should be surrounded by a strip, which should be connected with every edge and prominence, and be continued to the summit of each separate chimney. It is particularly necessary to guard the chimneys; for Mr. Morgan mentions a case, in which a house that had been guarded in most respects, according to the preceding directions, except that the chimneys were unprotected, was struck with lightning, which entered by one of the chimneys: here it spent its fury; but the chimney falling on the roof, did considerable damage. The principal objection to this method is the expense attending it; but this may be in a great measure avoided, by making proper use of the leaden pipes, gutters, and copings, which belong to most houses.

327. Before leaving the subject of conducting rods, we think it due to the ingenious professor Leslie of the University of Edinburgh, to notice an article on electrical theories, published by him in the Edinburgh Philosophical Journal for July, 1824. In this paper Mr. Leslie endeavours to show that thunder-rods are of no use whatever in the way of protecting buildings against the

effects of lightning. But lest we should in any degree misrepresent the professor's sentiments, by giving an abridged statement of them, we shall quote his own words :

328. But, whatever speculations we may form in regard to electrical light, and the mode in which the point and the knob produce their different effects, we must admit that the electricity is never communicated, in any perceptible degree, to a remote and unconnected body, but by means of a current of air; and this established principle will enable us to estimate the real effects of conductors or thunder-rods.

329. When two portions of air, near the point of saturation, and of different temperatures, are mixed, a quantity of the dissolved vapor is precipitated, and resumes its aqueous state. By this conversion the mass acquires electricity; and the consequent repulsion exerted, tends to disperse the minute globules of water, which will float in the atmosphere, or rather, will descend with that slow motion which is sufficient to occasion a resistance on their large surface equal to their gravitation. If the cloud thus generated reach the ground, it will soon communicate its electricity. If it be suspended at some height, the electrified air will stream from it in all directions; and if its formation be gradual, this discharge may suffice to waste its force. But when a vast cloud is suddenly formed, the aerial emission hardly impairs its electricity; and, as it is carried along, it continually approaches, by its attraction, to the surface, which assumes an opposite electricity; the air now rushes with violence, and the cloud bends faster downwards, till at last its lowest verge reaches the ground, and a total discharge is made. The magnitude of the stroke will evidently depend on the extent of the aqueous mass, the suddenness of its precipitation, and the rapidity of its descent.

330. The air, which streams in all directions from the cloud, is dissipated among the more remote portions, and thus gradually communicates its electricity. Hence, from the wide dispersion, Owing to the distance, the electricity of the air at the surface of the earth must be weak; and, even in the midst of the storm, the electrometer is less affected than if placed only a yard behind the prime conductor. Yet the action of the thunder-rod is confined entirely to the air which immediately surrounds it, and the quantity of aerial current which it can produce, must evidently be inferior to what is directed to the point, when held several feet from the conductor of an electrical machine. But, to avert the stroke, it would be necessary that the whole air between the surface and the cloud should be brought successively in contact with the top of the rod. Nor is this all; for the air will be constantly replaced by other electrified portions emitted from the cloud. The effect of the thunder-rod is therefore, comparatively, but a drop in the ocean. It may be easily shown that, however pointed and tapered, it would require 1000 years to guard at the distance of 100 yards; if terminated with a knob, it may take 10,000 years. Such are the raunted performances of thunder-rods, and such the advantages of their different forms! Nor

can we appeal to experience; it never can be proved that thunder-rods have produced peneficial effects, but several instances may be cited where they have afforded no sort of protection. Nay, we shall be convinced, that fully an equal proportion of the buildings armed with such supposed safeguards have been struck with lightning. But if thunder-rods are useless, they are also innocent; and, that they provoke the shaft of heaven, is the suggestion of superstition rather than of science. The cloud exerts an attraction, indeed, upon the surface of the ground, but the force depends solely on the distance, and is not, in the least degree, affected by the shape or quality of the substances below. It rolls towards the nearest and most elevated objects, and strikes indiscriminately a rock, a tree, or a spire.

331. If a thunder-rod be then a harmless, though idle, appendage to a house, why awaken uneasy apprehensions? It might at least inspire confidence in the moment of danger; and if happiness consists merely in idea, why not indulge delicious error?-Yet, though the inevitable stroke cannot be turned aside, its destructive effects may be lessened; and an investigation of the real action of thunder will conduct us to the proper principles.'-Ed. Phil. Jour. No. XXI. pp. 25. 28.

332. The above theory is certainly ingenious and worthy of attention, but it wants confirmation in another way than that attempted by its author. We would, with all deference to Mr. Leslie's well-known abilities, ask if he would advise large fires to be kept up round powder magazines for the purpose of preventing them from being struck by lightning? We do not know, either, why he advises that they should have copper conductors raised near them? nor why, to save ships, ribands of copper should be extended from the masts to the keel, since he affirms that the idea of diverting or dissipating the storm is wholly chimerical.

SPONTANEOUS ELECTRICITY OF THE AT

MOSPHERE.

333. Very numerous are the observations which have at different times been made by able observers on this branch of atmospherical electricity. We cannot here enter minutely into these, and shall therefore only offer an outline of their general results.

334. The earliest observations of this nature appear to have been those of Monnier; his experiments were made with an apparatus, which consisted of a pole thirty-two feet in height, insulated in a piece of turf, having at its top a strong glass tube, to which a tube of tinned iron was attached, and which terminated in a point. About the middle of this tube there was fastened a fine iron wire about fifty lines long, which, without touching any other body, was connected with a silk cord stretched horizontally. He found although the atmosphere was constantly electrified more or less, yet that in dry weather the electricity increased from sun-rise, when it was weakest, till about four o'clock in the afternoon, at which time it was strongest, gradually diminishing from that time till the dew began to fall, after which it diminished till midnight. The

sarae results were afterwards obtained by Saussure and Beccaria.

335. The abbé Mazeas made several observations with an atmospherical apparatus, consisting of an iron wire 370 feet long, raised about ninety feet from the ground, and properly insulated. The results of his experiments with this instrument were the following:-In very dry weather the wire readily attracted light bodies, if brought within three or four lines of it; and, if the weather was not stormy, the electricity of the air was about half as great as that of a stick of sealingwax two inches long. When he grasped the wire in his hand, the signs of electricity disappeared entirely, and did not return till after an interval of three or four minutes. He also found that the electricity of the atmosphere was not increased with storms and hurricanes unattended with rain; for during a violent storm of wind, which continued uninterruptedly for three days, in the month of July, he found it necessary to place the dust within four or five lines of the conductor, before it exhibited a sensible attraction. No change was produced by the different directions of the winds. In the driest nights of summer he never could observe any electricity in the air, but it began to appear in the morning at sunrise, and vanished in the evening at about half an hour after sun-set. In the month of July, in a very dry day, when the sky was serene, and the heat intense, he found the electricity stronger than he had ever observed it. The dust was then attracted at the distance of ten or twelve lines from the conductor.

336. Mr. Kinnersley made several observations on the electricity of the atmosphere in its ordinary state, and informs us that when the air was very dry, he found it always contained some electricity, and that he rendered this electricity visible by electrifying himself negatively, and holding in his hand a long sharp needle, which, in the dark, appeared luminous at the point.

337. Beccaria, who was particularly attentive to this subject, and made numerous experiments both with rods and kites, found that, during the three following states of the atmosphere, it afforded no sensible indications of electricity. (1.) When the weather was clear, but at the same time windy. (2.) When the sky was covered with well defined black clouds, moving slowly. (3.) During moist weather, when not actually raining. The electricity was always perceptible when the sky was clear and the weather calm; in rainy weather, and when it did not lighten, the electricity appeared always a little before the rain fell, continued during the falling of the rain, and disappeared shortly after the rain ceased. The higher he raised his rods, the stronger he found these electrical signs.

338. M. de Saussure, however, has furnished us with the most extensive course of experiments on atmospherical electricity. He informs us that it is constantly varying, according to situation. It is generally strongest in elevated and insulated situations, not to be observed under trees, in streets, in houses, or any enclosed places, though t is sometimes to be found pretty strong on quays and bridges. It is also not so much affected by the absolute height of the places as

their situation; thus a projecting angle of a high hill will often exhibit a stronger electricity than the plain at the top of the hill, as there are fewer points than in the former to deprive the air of its electricity. This variation in the intensity of the electricity of the atmosphere, seems influenced by numerous circumstances, some of which it is difficult satisfactorily to account,for.

339. When the weather is not serene, it is impossible to assign any rule for their variation, as no regular correspondence can then be perceived with the different hours of the day, nor with the various modifications of the air. The reason is evident; when contrary and variable winds reign at different heights, when clouds are rolling over clouds, these winds and clouds, which we cannot perceive by any exterior sign, do, nevertheless, influence the strata of air in which we make our experiments, and produce these changes of which we only see the result, without being able to assign either the cause or its relation. Thus, in stormy weather, we find the electricity strong, then completely gone, and in a moment after rise to its former force; one instant positive, the next negative, without being able to assign any reason for these changes.

340. M. Saussure says, that he had known these changes succeed with such rapidity, that he had not time to note them down. When rain falls without a storm these changes are not so sudden; they are, however, very irregular, particularly with respect to intensity of force; the quality of them, too, is more constant. Rain or snow almost uniformly gives positive electricity. In cloudy weather, without rain or storms, the electricity follows generally the same laws as in serene weather. Strong winds generally diminish its intensity; they seem to mingle together the different strata of the atmosphere, and make them pass successively towards the ground, and thus distribute the electricity uniformly between the earth and the air. M. Saussure has observed a strong electricity with a strong north wind. The state of the air in which the electricity is strongest is foggy weather; this is also accompanied with electricity, except when the fog is going to resolve into rain.

341. The most interesting observations, and those which throw the greatest light upon the various modifications of electricity in our atmosphere, are those made in serene weather. In winter, and in serene weather, the electricity is generally weakest in the evening, when the dew has fallen, until sun-rising; its intensity afterwards augments by degrees, sometimes sooner and sometimes later; but generally before noon, it attains a certain maximum, whence it again declines, till the fall of the dew, when it will be sometimes stronger than it had been during the whole day; after which, it will again gradually diminish during the whole night; but, it is never quite destroyed, if the weather is perfectly serene. Atmospherical clectricity seems, therefore, like the sea, to be subject to a flux and reflux, which causes it to increase and diminish twice in twenty-four hours. The times of its greatest force are some hours after the rising and setting of the sun : those when it is weakest precede these periods.