increasing the number of the groups. The first method would have been cumbersome; but for a reason which I shall explain, I did not expect that it would have the desired effect. However, even for the verification of my conjecture, I made the following experiment.

Exp. 29.-I procured 10 tinned iron plates, 4 inches square; 10 others round, of only 0.5 inch diameter; and I took 10 of my plates of 1.6 inch diameter. I cut pieces of Dutch giit paper, the size of each of them, and compared their respective effects on my condenser. The result was beyond my expectation; I had only conjectured that the increase of size would not increase the divergence of the gold leaves; now this was not only verified, but the largest plates produced the smallest effect. However, this unexpected difference probably proceeded from some accidental cause, which I had no time to investigate, and I considered this experiment, which I have often repeated under various forms, only as ascertaining the following proposition: that for the electric motions, considered solely with. respect to the quantity of divergence in the electroscope, the size of the plates is indifferent; though, for the frequency of the strikings of the little pendula, and the intensity of the effects when the extremities of the pile are connected together, with the same number of groups, these effects increase with the size of the plates. This distinction, to which I shall return in the following paper, constitutes a part of the theory, which, as it made me foresee the result of the above experiment, I shall now explain.

I am indebted for the ground of this theory to Sig. Volta, who, when in 1782, he showed me, at Paris, his then new-invented admirable condenser, explained to me, that it could not serve to mauifest minute degrees of deviation from the electric standard, when belonging to small bodies, but only to bodies of such an extent or nature, that the application of the condenser (by this taking its share of that deviation) does not sensibly lessen it. As an example of the first case, he gave me the atmosphere; and with respect to the nature of bodies, he took a Leyden phial discharged without a continued contact, the residuum of which, from its nature, may affect a moderate sized condenser without being much lessened and to show me the necessity of this condition, he made use of the following analogy. When a piece of ground, by being swampy, indicates some stagnant water, if a well be dug there, the water will not fill it up to the level of the stratum of earth whence it proceeds, unless this stratum be of such an extent, that the quantity of water which gathers in the well has no sensible proportion with that contained in the stratum; so that the subtraction of this quantity cannot affect the level at which the water stands in the stratum. We have also an example of this case in the subject of electricity: when an insulated electrified body is small, we cannot know its real degree of electrification by applying a common electroscope; because this, sharing the deviation of the state of the body from the electric standard, lessens it too much for expressing what it was before that application.

I shall use the former of these examples in explaining my theory concerning the difference of effects of the size of the plates, and the number of the groups, according to the use of the pile; and this explanation will chiefly consist in fixing the points of analogy between the two objects.

1. I compare the number of groups in the pile to the elevation of a stratum whence water issues into a well.

2. The size of the plates, to the extent of this stratum.

3. The degree of divergence in the electroscopes at the extremities of the pile to the level which the water can attain in the well without over flowing.

These first analogies are sufficient to explain the case of the last experiment. When we attend to what is directly expressed by our electroscopes, we certainly do not expect that this instrument shall indicate the quantity of electric fluid possessed by the bodies to which it is applied; for this would require also to measure their surface; we expect only to know the comparative density of the electric fluid among bodies, or its power to produce certain degrees of divergence in the electroscope we use.* Such therefore are the indications of the electroscopes at the extremities of the pile: they express certain degrees of density of the electric fluid on them, which are the same whatever be the area of the plates; these degrees depending only on the number of the groups, because each group contributes to increase the density of the fluid on one of the extremities by lessening it on the other. Thus it is that the divergence, both in plus and minus, of the electroscopes, at the extremities of the pile, is proportional only to the number of groups, in the same manner that the height at which water stands in a well is proportional only to the elevation of the stratum whence it proceeds.

4. But when, in order to produce a current of water, a pipe is placed, or a trench is cut, on the side of a well, below the level at which the water stood in it, the current will be greater and more permanent, in proportion to the extent of the stratum, of the same elevation, whence the water proceeds; and also, when the divergence

I shall take this opportunity of explaining why I use the word electroscope, and not that of electrometer; it is because there is no instrument entitled to the latter denomination, at least admitted among experimental philosophers. Indeed, of our instruments serving to measure the degrees of intensity of physical causes, I know none absolute, except hygrometers, such as have been constructed by M. de Saussure and myself; for though these instruments are made of different substances, and differ in some other respects, we have obtained in both an absolute zero, and absolute maxima in determined cases, as well as determined degrees of intensity, of their object, namely moisture: all the other instruments intended for the same purpose, to my knowledge, are only hygroscopes, indicating variations in moisture, without determined points, or degrees common to them. Thus no kind of physical instrument has yet obtained the conditions of an absolute measure, but the above hygrometers, so little thought of by experimental philosophers, though very important in meteorology. The thermometer has obtained two fixed degrees of heat, and determined divisions of the interval between them,

of the little pendula of the electroscope exceeds the extent that it can have without one of them striking the side, then falling by a momentary contact with the tin foil, which communicates with the ground, it will sooner rise and strike again, with the same number of groups, in proportion to the size of the plates; which last circumstance increases also the current of electric fluid circulating in a pile, the extremities of which are connected together by a conductor.

5. The water of all springs has the same source, namely, the rain water percolating through the ground, and retained on some impervious stratum, either argillaceous or stony. If this water do not find in its way any substance with which it can combine, it comes out as it had fallen on the ground: but if in its course it combines with any substance, it may come out with certain chemical properties, different according to the substances which have combined with it.

The case is the same with respect to the electric fluid which pervades the pile: its source is no other than the electric fluid ditfused over all terrestrial bodies, therefore over the pile itself. However we should be ignorant of the constant existence of this fluid, over us and around us, were it not that, by artificial or natural operations, its density may either be increased or diminished on insulated bodies: this is the only circumstance which makes it appear, and that by the electroscope alone; for as long as this fluid remains in a state of equal diffusion over all bodies, it is manifested by no effect hitherto discovered. The friction between two bodies disturbs that equilibrium, in a manner which I shall show in a future paper on the Analysis of the Electric Machine. But in the pile, which is my present object, it is by a property of its composition, that the equilibrium of the electric fluid is disturbed, whence proceed either the motions of the electroscopes, or a circulation of the fluid through the pile, when the extremities of the latter are connected together by a conductor. Now, in the last of these cases, if the electric fluid, in its course meets with no substance that changes its state; as is the case in a pile composed of tinned iron, or zinc plates, separated by Dutch gilt paper; we are indeed informed by the electroscopes of its accumulation on one extremity of the pile, and

by which means experimental philosophers understand one another when they indicate certain degrees of heat. I have also constructed an electrometer, which possesses the same conditions with respect to degrees of electrification, which is described in my work, Idées sur la Métérologie; but not having been attended to by experimental philosophers, I have not been induced to follow the extension of this measure down to the minute degrees of intensity indicated by the gold leaf electroscope, as I could not expect that it should be more noticed therefore admirable as is this instrument for its sensibility, it

affords us no comparable measure. In this imperfect state, however, there is, in every electroscope, a property which belongs to no other physical measure, namely, a natural and absolute standard of plus and minus, which is constant, as to its general determination, and is the actual electric state of the ambient air, or the ground; though variable as to the absolute quantity, as are these electric states; which difference will be one of the objects of this paper.

its deficiency on the other; however, neither chemical effects in the circuit, nor the shock, are produced; because the fluid remains unaltered but when it pervades a pile, wherein, by a liquid being placed between the two metals, there is calcination of one or both of the latter, new effects appear: if the liquid be pure water, chemical effects are produced in the circuit, but there is no shock; if it be an acid, both effects are produced.

These experiments, especially on the different effects of the number of the groups, and of the size of the plates, with the above theory on the cause of their different effects, were contained in my paper delivered to the Royal Society the 30th of May, 1808, about one month before Mr. J. G. Children executed in presence of Mr. Davy and Mr. Allen the grand experiment of the same kind related in Part I of Ph. Trans. for 1809, by which the theory which I had already announced was confirmed.

But here two questions arise, which go deeper into the mode of action of the galvanic pile, and they are these: 1. Of what nature is the modification produced in the electric fluid, when it pervades a pile wherein the calcination of some metal is going on? 2. What is the cause of the motion of this fluid in the pile, whether producing, or not producing the shock and chemical effects in the circuit?

The solution of the former of these questions, which leads to that of the latter, depends on the nature of the electric fluid; a subject much too long to be treated here; but it is fully detailed in both the works I have already referred to ;* I shall therefore here confine myself to the conclusions contained in these works, as deduced from uninterrupted series of experiments, of which I shall only detail the part necessary to my subject.

None of the phenomena observed in our common electrical experiments, namely, the charge and discharge of the Leyden vial, the electric motions, the effects of the clectrophorus and of the condenser, had been really explained, till the inventor of the last two instruments, Sig. Volta, had formed his theory on the electric influences, which threw the first true light on the modifications of the electric fluid; and which, in the course of various experiments I made to follow it through all the electric phenomena, gave rise to the system on the nature of the electric fluid, which I shall here briefly state.

This fluid, far from being a simple substance, is an astonishing compound and first, in its state which may be called natural, that, I mean, in which it is diffused over all bodies, it is found composed of two main parts, from which all the above-mentioned phenomena arise. One of these two constituent ingredients of the electric fluid in this state is a substance, which, by itself, is not expansible (as in steam, also an expansible fluid, there is a substance which is not expansible by itself, namely water); this substance in the eletric

Idées sur la Météorologie, and Traité élémentaire sur le Fluide électro-gal vanique.

fluid I have called electric matter; and its function, which I shall soon point out, is very distinct. The other ingredient is an excessively subtle fluid, which (as fire in steam) uniting with the non-expansible substance, produces the expansibility of the aggregate. In my French works I have called the latter fluide déférent; but here I shall call it vector, a short word of the same import, signifying that it carries along the electric matter (as, in steam, fire is the vector of water).

The electric vector instantly pervades all bodies, and carries the electric matter through conductors, but not through non-conductors, such as glass and resinous substances: when a current of electric fluid arrives on one side of a lamina of these substances, and its vector, in order to establish its own equilibrium beyond it, pervades the lamina, it deposits the electric matter on the surface of the latter, where it remains adherent, till a current of vector pervades the lamina in the opposite direction, or it is taken up slowly by the vector in the air (as fire in steam, when it pervades a glass lamina to establish its own equilibrium beyond it, deposits the water on the side which receives the steam, where it remains, till it is carried away, either by fire coming from without, or by that spread in the air).

I come to the peculiar function of the electric matter in the above indicated phenomena: it is the sole cause of electric motions, resulting from a greater or less proportional quantity of it, than is possessed by the ambient air; to which subject I shall return: the vector has no share in these motions, but as the vehicle of the electric matter acting in their phenomena. (As, with regard to steam, it is only water that produces the hygroscopic phenomena, without any interference of fire, except as the vehicle of water).

By this system of a first composition of the electric fluid, the phenomena, which I have introduced in the beginning, are clearly explained in all their modifications, as I have abundantly proved by direct experiments in my works. But as long as the electric fluid remains in what I have called its natural state, moving along conductors and fixed on non-conductors, it produces no chemical effect hitherto known: what then does happen, when it produces these phenomena ?

If we attend to this change, we shall observe a circumstance sine qua non, which is to contain some cause; it is, that the conductor, along which the electric fluid moves, must be interrupted. Now, when in this case the electric fluid darts through the air, three new phenomena are observed, lucidity, heat, and a particular odour. This cannot but indicate the decomposition of some particles of the fluid, occasioned by an excess of density, from which light, fire, and an odorute substance are disengaged: as when steam (to which from the beginning I have compared by analogy this system on the nature of the electric fluid) becomes too dense for the actual temperature, some of its particles, being decomposed, emit water and fire.

These new substances, light, fire, and an odorate substance, thus manifested in the composition of the electric fluid, are neither the