it emanates, seeming to issue from their interior. This is the most common of all the forms.

Those flashes of light unattended by thunder which illumine the horizon for hours, at times, on summer evenings, have been characterized as heat lightning, and are generally attributed to reflection from the atmosphere of the lightning of clouds so distant that the thunder cannot be heard. Sometimes this light is diffused over the entire heavens, the electricity of the clouds escaping in flashes too feeble to produce audible sound. Especially may this occur when the air is moist, so that it conducts electricity with tolerable facility, and yet resists its passage sufficiently to develop a feeble light. It may here be remarked that the form under which lightning is presented, whether zigzag, sheet or heat lightning, depends upon the position of the observer. That which may present itself as zigzag lightning to one observer, may appear as sheet or heat lightning to another located differently.

While the first form of lightning is that which will principally engage our attention, yet, lightnings of the second class are so unique in their manifestations, I can hardly resist the temptation to draw examples of them from that part of the admirable meteorological essays of Francois Arago which treats of thunder and lightning.* Unlike the instantaneous flashes of the other. forms, lightnings of this class are visible for one, two, or even ten seconds of time. Their movement in descending from the clouds to the earth is so slow that their march can be followed by the eye and their rate be estimated. They occupy definite spaces, are of globular form, and are seemingly balls of fire. Against them lightning conductors and all other means of protection which man has devised, seem to be essentially powerless.

We give below a few from the long list of examples referred to above:

"At Couesnon, near Brest, among the ruins of a church which had been entirely destroyed, different witnesses agreed in attributing the catastrophe to three fiery globes, each three or four feet in diameter which united, and then proceeded with a very rapid course in the direction of the church.'"

In January, 1770, a thunderbolt fell on the tower of Schemnitz in Hungary. Its form was that of a globe, and its size as large as a cask.'

* While facts have been drawn from any available sources, I desire to acknowledge particular indebtedness to the work referred to above.

"On the 14th of February, 1809, the ship of the line, 'Warren Hastings,' which had only been launched a few days before at Portsmouth, (Eng.) was struck three times in a very short space of time. On each occasion the lightning approached the mast under the form of a ball of fire."

"A short time after Philip V. had made his entry into Madrid, the palace was struck by lightning. The persons assembled at the moment in the royal chapel saw two balls of fire enter it. One of these balls divided into several smaller ones, which before disappearing, bounded repeatedly like an elastic ball."

"On the 20th of June, 1772, while a thunder storm rolled over the parish of Steeple Ashton in Wiltshire, a globe of fire was seen to hover in the air above the village for a considerable time, and afterwards to fall perpendicularly upon the houses, where it did much damage."

At the same place, and on the same day, the reverend Messrs. Wainhouse and Pitcairn, who were in a room in the parsonage, suddenly saw appear at the height of their faces and at about a foot from them, a globe of fire of the size of a fist. It was surrounded by black smoke. In exploding it made a noise which might be compared to the discharge of several pieces of ordnance at once. Immediately afterwards a strongly sulphurous vapor spread throughout the house. Mr. Pitcairn was dangerously wounded; his body, clothes, shoes and watch, presented all the same appearances as those attendant on a stroke of lightning of the more usual kind. Different colored lights filled the apartment, and were violently agitated to and fro."

Professor Richmann of St. Petersburg, in 1752, was instantly 'killed while performing the experiment of withdrawing lightning from clouds. The engraver Solokoff who was present, declared that the lightning which killed the physicist had a globular form.

"In 1809, lightning entered the house of Mr. David Sutton at Newcastle-on-Tyne, through the chimney. After the explosion, several persons saw on the floor, at the door of the drawing-room in which they were assembled, a globe of fire which remained stationary; it afterwards advanced into the midst of the room and broke into several fragments, which exploded in their turn like the stars of a rocket."

M. Babinet communicated to the Academy of Sciences (France) on the 5th of July, 1852, the following note:

"The object of the present notice is to bring before the

Academy a case of globular lightning, which the Academy had charged me a few years ago, (June, 1843,) with the care of investigating and authenticating, and in which the ball of lightning had struck a house (Rue St. Jacques in the neighborhood of the Val de Grace,) as it withdrew. The following is a brief summary of the account given by a workman into whose room the globular thunderbolt descended and then remounted:

After a rather loud thunderclap, but not immediately after it, the workman, a sailor by trade, being seated by his table finishing his meal, suddenly saw the chimney-board fall down, as if overset by a slight gust of wind, and a globe of fire the size of a child's head come out quietly from the chimney and move slowly about the room at a small height above the tiles of the floor. The sailor said it looked like a good sized kitten rolled up in a ball and moving without showing its paws. It was bright and shining, rather than hot and burning; the man said he felt no sensation of heat. The globe came near his feet like a young cat that wants to play and rub itself against its master's legs; but by moving his feet aside and making various precautionary manoeuvres,-all done by his own account very gently, he avoided the contact. It appears to have played several seconds about the feet of the workman, who remained seated, his body bent over it and examining it attentively. After having tried some excursions in different directions, but without leaving the middle of the room, it rose vertically to the height of the man's head; to avoid its touching his face he raised his body and threw himself back in his chair, still keeping the meteor in view. When it had risen, three or four feet above the tiled floor, the globe became a little elongated, and rising obliquely directed itself towards a hole pierced in.the chimney three and a half feet above the mantleshelf. The hole had been made to allow a stove pipe, which the workman used in winter, to pass through, but according to his own expression, the thunder could not see the hole, for it was covered with paper which had been pasted over it.' The globe of fire however went straight to the aperture, unpasted the paper without hurting it, and made its way into the chimney; then when it had just had time at the pace it was going, that is to say, pretty slowly to get to the top of the chimney (at least 20 metres, or 66 feet from the ground of the court yard) it made a dreadful explosion, which destroyed the upper part of the chimney, and threw the fragments into the yard on the roofs of smaller buildings which

they broke through; happily no one was hurt. The movements of the luminous globe were always slow, and not by jerks. Though bright it was not dazzling, and no sensible, heat came from it. It does not appear to have had a tendency to follow conducting bodies, or to have been impelled by currents of air."

A sufficient number of examples have been already adduced to give an approximate idea of the nature of these fiery globular masses. While they are not of frequent occurrence, yet they have been seen sufficiently often, and studied with so much care, that their existence as a peculiar form of lightning is abundantly attested. Their cause has been variously assigned. Professor Loomis says, they probably result from." a charge of electricity unusually intense, which forces a direct instead of a circuitous passage through the air." By some they are thought to be agglomerations of ponderable substances in a state of great tenuity, and strongly charged with electricity.

Whatever may be their composition or their cause, in the slowness, uncertainty and peculiar character of their movements, and in the extent of the damage, resulting from their explosion, they rank among the most marvellous objects presented in the whole range of meteorology.

That form of lightning with which we are most deeply concerned presents a long, irregular jagged line of light, resembling the spark drawn from an electric machine. Its zigzag path is regarded as due to the compression of air before the electric fluid, by which greater resistance is produced. When its course is strongly resisted in one direction, it turns aside following the line of least resistance, until the air is again compressed before. it, when it turns again as before. The duration of an ordinary flash of lightning as determined by receiving the light of an electric discharge upon a white disc marked with black rays radiating from the centre, the disc being made to revolve with great rapidity, has been repeatedly determined to be less than a thousandth part of a second of time. The length of the zigzag path ranges from short spaces to the distance of ten miles. Thunder clouds have been observed of all heights from less than one-fourth of a mile to at least three or four miles.

Inasmuch as low clouds come in contact with the sides of mountains, and even of high hills, the following interesting inquiry is suggested. Can clouds from which lightnings are incessantly darting be traversed without imminent danger?

However great the peril may be such clouds have been traversed without disastrous results. From many instances on record, I select the following:

In August, 1750, an ascent was made of the little mountain Boyer in France. Three-fourths the way up the mountain was a stationary cloud, from which thunder was from time to time heard. From the moment the cloud was entered, "the thunder no longer manifested itself by sudden claps, alternating with intervals of silence; it now made a continual rumbling, resembling that of a heap of walnuts rolled upon the floor. When the observer had attained the summit of the mountain he found himself above the cloud which had not ceased to be a thunder cloud, for it was traversed by brilliant lightnings, and loud detonations issued from it."

Observers upon the Pyrenees in the midst of thunder clouds have noticed that their hair and the tassels of their caps stood on end, and that a hissing sound was produced from prominent points. A position above a thunder cloud cannot be regarded as unquestionably safe. Lightning has been known to issue from the upper surfaces of clouds, and strike objects upon the summit of mountains above them. Thus on May 1st, 1700, seven persons were killed in a church on the summit of Mount St. Ursula in Styria.

The thunder which ordinarily accompanies a discharge of lightning, is regarded as resulting from the collision of particles of the atmosphere as they reënter the partial or total vacuum produced by the lightning in its swift passage through the air. As is well understood the interval between the flash, and the report indicates approximately the distance of the cloud, or rather of that part of the cloud from which the sound which first reaches the ear, emanates, an allowance of about five seconds for the distance of a mile being required. The largest interval mentioned by any observer is 72 seconds, indicating a distance of nearly 15 miles. The next longest interval recorded is 50 seconds, corresponding to a distance of about ten miles. It seems remarkable that the sound of thunder has not more of a diffusive character. The sound of cannon may be heard to a much greater distance.

The average interval between flash and report is 12 seconds, and the shortest interval noted less than one-half a second. The average duration of peals of thunder is 22 seconds, and the longest duration on record is 56 seconds. The prolonged sound in