lanced by the friction, and when the motion was rendered so slow as seven feet seven inches in seven seconds, it was very little accelerated; and in this state, the great cylinder being charged, the model was suffered to pass; and though the velocity was less than three-quarters of a mile in an hour, the point was struck.

Exp. 32. When the great ball of the less substitute was placed at the greatest striking distance from the ball at the end of the great cylinder, I fixed a needle into the under side of the remote end of the less substitute, with the point downwards; opposite to this point, and on the same stand described in the twenty-second experiment, was fixed another needle; so that the two points were opposed to each other. The space between them was varied from time to time, in order to find the greatest distance at which the lower one could be struck. On charging the great cylinder, it appeared that the greatest distance in this case was five inches and one-quarter.—Exp. 33. Upon repeating this experiment, whilst every circumstance remained the same, excepting that instead of the point below, a rounded end was put in its place, and after charging the cylinder again, it appeared that the greatest distance at which the lightning struck the rounded end, was not more than two inches and three-quarters. And the largeness of the spark, and the loudness of the explosion appeared to be less considerable than in the thirty-second experiment.

During the course of this inquiry, having occasion to try some experiments in the dark, I observed a curious circumstance, which seemed to show, that a point had a far greater influence on the charged substitute, in certain circumstances, than a rounded end had when it was placed in the same situation.

Exp. 34. The circumstance alluded to, was an appearance of light on the brass ball that was fixed at the end of the great cylinder, when the copper ball of the less substitute was opposed to it at the greatest striking distance, as in the eighteenth experiment, every other circumstance remaining the same; and while the model with its pointed conductor, stood on the table, directly under the tin ball, fixed at the remote end of the less substitute: soon after seven or eight turns of the wheel, a light began to appear on the brass ball, and continued to increase in brightness till the moment it burst forth in an explosion towards the copper ball. The part of the brass on which the light appeared was that next to the copper ball, and the general appearance of it was round, and sometimes more than half an inch in diameter.-Exp. 35. On repeating this experiment with a rounded end instead of a point, and at the same distance from the tin ball, though every other circumstance continued the same, there was no such appearance.-Exp. 36. But when the rounded end was moved considerably nearer, that is, within sixtenths of an inch, a light was visible; but then it was faint, and not more than one-tenth of an inch in diameter, even at the instant before the explosion happened.

Thirteenth Observation.-By the first of these experiments it appears, that the influence which the point had on the whole of the fluid

contained in the great cylinder, was such as to cause a general tendency of it towards the less substitute; but on account of the resistance which seeined to operate at the surface of the brass ball, it was there stopped, and by degrees accumulated, till such time as the accumulation was great enough to overcome that resistance. Now according to this manner of reasoning, the point did not draw the fluid out of the great cylinder silently; but when the accumulation had got to a sufficient degree, a sudden explosion ensued, more or less violent, according to the circumstances which accompanied the experiment.

Fourteenth Observation.-From the other experiment it appears, that the rounded end had not so great an influence as the point on the charge in the cylinder; because we were obliged to bring it five times nearer before any light could be perceived at all; and even then it was so faint and inconsiderable in its diameter, compared with the other light produced by the influence of the point, that it manifestly confirmed the truth of the last observation. I shall now proceed to make some general deductions from what has been already related.

It seems to be clear then, that in all experiments made with pointed and rounded conductors, the rounded ones are by far the safer of the two, whether the lightning proceed from one or more clouds; that those are still more safe, which (instead of being as Dr. Franklin recommends, ten feet high), are very little, if at all, above the highest part of the building itself; and that this safety arises from the greater resistance exerted at the larger surface. The luminous appearance at the end of the brass ball, occasioned by the point in the thirty-fourth experiment, manifestly showed that there was an accumulation of the fluid within that part of the ball, in consequence of some resistance: for when the resistance at the surface of the brass ball was at last overcome by the influence the point had on the charge, the explosion took place immediately; and that, not only between the two substitutes, but also between the end of the less substitute and the point. A cloud, therefore, that happens to be charged, and within the striking distance of another cloud which is not charged, and also equally within the influence of a pointed conductor, must necessarily produce similiar effects with those mentioned in the thirty-fourth experiment. On the other hand, clouds that are circumstanced like those above, and not within the influence of a rounded conductor, will pass quietly over such a termination, and without any explosion.

On Acceleration, and its Effects. From considering the extraordinary effects which have sometimes been produced on gross matter by lightning, and the distance there frequently is between thunder clouds and the earth, when such effects take place, I suspected that those effects might in some degree be owing to an increase of the velocity of the fluid which produced them. To try whether this really was so, it seemed necessary to have an apparatus

of a far greater length than the great cylinder; I therefore made use occasionally of the long wire which has been already described.

Exp. 37. On connecting one end of this long wire with the further end of the great cylinder, and the other with one end of the brass drums, I found that about six uniform turus of the wheels, with a moderate velocity, were required to cause the appearance of a small stream of light at the top of the spout described in the first experiment; when the model, with the pointed conductor on it, stood directly under the great cylinder, but at the distance of five inches. Exp. 38. When the great cylinder was unconnected with the long wire and brass drums, and while the model, with the same conductor on it, remained in its place, about two turns, with the same velocity, were sufficient to charge the great cylinder, so as to cause a similar appearance at the spout.-Exp. 39. On separating the great cylinder from a fourteenth part of it, the model and machine continuing in their places, it was found that half a turn of the wheel was sufficient to charge the little cylinder, so as to cause the like appearance at the spout.

Fifteenth Observation.-Now these differences in the numbers of turns required for causing similar appearances, when the several charges were given in the thirty-seventh and thirty-eighth experiments, could not arise from a difference in the quantity of metallic matter contained in the respective substitutes; because the tin-foil which covered the great cylinder was found to be nearly three times heavier than the weight of the whole wire. Neither could these differences be owing to a difference in the quantity of surface of the respective substitutes; because the surface of the great cylinder was found to be ten times greater than the surface of the wire. Those several differences must therefore depend on some other cause; and, as a true knowledge of this cause may be of some moment in the present inquiry, we must endeavour to find it out by experiment and observation.

Exp. 40. When the great cylinder, with the wire and brass drums, were charged with a very small quantity of this fluid, by the wheel being turned something less than a quarter round, there was the moment after, a visible explosion, and a sensible effect perceived at the remote end of the wire. When half a turn was given, these effects were greater; and, after a whole turn, the quantity of the fluid accumulated in this great apparatus was increased considerably.

Sixteenth Observation.-Now, something must have hindered the fluid from getting out of the cylinder and wire all the time they were charging, otherwise we should not have been able to have caused the least accumulation; for, from the nature of this fluid, there cannot be any accumulation without some resistance to occasion it. And whatever the nature of that resistance may be, experiments show, that there are certain bounds prescribed to its power of acting, and which in particular circumstances seem to be very easily surmounted.

Exp. 41. When the great cylinder and wire with the drums

were fully charged, and a person, standing on the wire which communicated with the well, suddenly approached the brass drums with his hand, an explosion ensued, which indeed was neither so large, nor did it take place at so great a distance, as might have been expected; yet the person received a violent sensation, not unlike that produced by the Leyden phial, as it affected his body quite through, from the hand that took the discharge to the feet that stood on the wire. Exp. 42. On repeating the above experiment, with the great cylinder only, and when it was fully charged, the explosion appeared stronger, and the distance it struck at greater, than in the other case; and yet the sensation received was not near so violent as when the long wire was connected with it.-Exp. 43. When the little cylinder by itself was fully charged, the effects were very inconsiderable, compared with those from the great cylinder: for, in this case, the person standing on the wire of communication was affected in his hand only, and that no farther than the wrist.

Seventeenth Observation.-When all the circumstances in the last two experiments are considered, we may safely conclude, that a difference in sensation, produced by the two cylinders, could arise from no other cause than a difference in their lengths; the one being fourteen feet longer than the other, and both in other respects nearly similar; since the sensation perceived in the thirty-eighth experiment, where the long wire was employed, was considerably greater than when the great cylinder alone was charged, we seem to have sufficient reason to apprehend that the effects of every charge, as to sensation, will be proportional to the length of the body charged; provided the charge be uniform from end to end in every experiment. Apprehending that, if some of the circumstances employed in producing the charge were varied, we might possibly obtain a greater charge than we had yet found, I made the following experiments.

Exp. 44. Instead of one machine to charge the great apparatus, I made use of two. The glass cylinders belonging to each were of the same length and diameter nearly. One of those machines was continued in its usual place, which was not far from the near end of the great cylinder. The other stood at the farther end of the brass drums. After connecting the long wire with the great cylinder and brass drums, in the manner before described, the wheels of both machines were put into motion, with equal and uniform velocities and after six turns of each wheel, and waiting above eight seconds, a person suddenly approached the brass drums with his band; immediately an explosion took place, and a disagreeable sensation was perceived. The discharge was then made at the nearer end of the great cylinder, and there seemed to be no difference in the effect.

Exp. 45. On repeating the experiment with one machine only, and after the same number of turns of the wheel with the same velocity, and waiting above eight seconds also, the same person suddenly caused an explosion with the same hand. But the sensation in

consequence of it was very different from the last experiment; for he declared it was little more than half so violent.

Exp. 46. I now charged the long wire only and fully, and with one machine: the explosion, in this case, appeared not very large, but of a reddish hue; and the distance it struck at was not more than an inch and a half; however, the sensation across the body was at that instant sharp and violent, but not quite so disagreeable as when the great cylinder was connected with it, and similarly charged.

Exp. 47. Having procured an equal quantity of the same kind of wire, and of the same diameter, with that which was suspended and tried in the last experiment, it was placed in the form of coils on a board, fixed on the top of a long stand of glass, without having any connection with the great apparatus. These coils were then fully charged by the power of one of those machines only. The sensation they afforded, in consequence of causing sparks, was very inconsiderable, compared with what had been observed in the last experiment.

Exp. 48. The several coils of wire employed in the last experiment, and seven hundred yards more in coils also, were joined together at their several ends. These coils being then strung on silk lines, were drawn out into a form resembling that of a screw, and separated from each other in such a manner all along as to occupy one hundred yards of silk line. The several diameters of these coils, at a mean, were about fifteen inches. As I had so short a time in which to prepare and suspend them properly, the disadvantage of their touching and intersecting each other in many places could not be prevented; however I found that the sensation caused, after charging this wire, was nearly equal to that which had been experienced from the long wire in the forty-sixth experiment.

Exp. 49. On joining the farther end of these coils to one end of the long wire, so that the whole length was in this experiment about three thousand nine hundred yards, and afterwards charging the nearer end of the coils, and without the great cylinder, the sensation complained of by two indifferent persons, was twice as violent as the sensation perceived by the same persons when the long wire alone was charged.

It may be now proper to make some general observations respecting the explosion itself, and the quantity of the fluid discharged in consequence of it. After many experiments we found, that when the great apparatus was fully charged, and the motion of the wheel suddenly stopped, it appeared, that a single explosion at either end of it, instantly discharged the contained fluid; but never so effectually as to leave no remainder: for the quantity which did remain was generally sufficient to cause a second explosion perceptible to the sense of feeling, as well as to that of sight. Now before the great explosion was caused, the fluid accumulated in the apparatus must have been diffused equally through it, in consequence of its elastic principle; and being so circumstanced, a sudden application of