separated into pairs by some conducting substance that does not interfere with their electromotive powers. To ascertain if a liquid was essential to this effect, he mounted a pile with pieces of cloth not moistened, and he found the electric effects were still produced, but somewhat weaker than with the wetted cloth. He then instituted a series of experiments, successively mounting the pile with different animal and vegetable substances, interposed between the pairs of metal, instead of wetted cloth. Of the various substances tried, he preferred writing paper, as the most convenient of those that were efficient. The apparatus constructed in this way was found to have the same electrical indications as the common Voltaic pile, but it produced no che mical effects, however numerous the pairs of plates; nor was any oxidation of the zinc produced by its most protracted action. These circumstances led to the idea, that, by the extension of the number of groups, a kind of perpetual electric machine might be formed; and, as in the previous trials, it had been found that the effect was rather increased by pasting the paper upon the silver or copper. Dutch gilt paper, which consists of thin copper leaf, laid upon paper, was employed instead of the usual silver, or copper plates, and moist conductors: 800 plates of tinned iron being put together with the same number of Dutch gilt paper between them, the copper sides being all turned in one direction, the combination was found to affect the electrometer more powerfully than any Voltaic battery had been ever observed to do; but on the application of the usual glass tube with water, no chemical effect was noticed. The apparatus was left for a considerable time, and its action on the electrometer continued without diminution; and subsequent experience has shown that it does so for any period during which the experiment has been continued. Thus was invented a new and important Voltaic arrangement, highly valuable both in a theoretical and practical view: in the former, as separating the pure electrical effects of the Voltaic battery from its chemical power, and demonstrating the permanence of its electro-motive faculty in the latter, as providing a spontaneous and permanent electrical machine, in which the opposite electrical states perpetually exist, without any new excitement. Besides these properties, the new apparatus promises to become an important meteorological instrument; for the degrees of its electrical indications have been observed to vary with the different seasons of the year, and are probably influenced by some of the causes by which our atmospherical phenomena are produced.

122. To distinguish this instrument from the usual Voltaic apparatus, from which it differs in many respects, M. De Luc proposed to call it the Electric Column,' an appellation sufficiently appropriate, since the effects it produces are purely electrical.

123. Mr. Singer made very numerous experiments, on the constructions of such columns, and varied their combinations most extensively. The materials he preferred, are thin plates of flatted zinc alternated with writing or smooth

cartridge paper, and silver leaf. The silver leaf is first laid on paper, so as to form silvered paper, which is afterwards cut into small round plates by means of a hollow punch. In the same way an equal number of plates are cut from thin flatted zinc, and from cominon writing or cartridge paper. These plates are then arranged in the order of zinc, paper, silvered paper with the silvered side upwards; zinc upon this silver, then paper, and again silvered paper, with the silver side upwards; and so on, the silver being in contact with zinc throughout, and each pair of zinc and silver plates separated by two dises of paper from the next pair. An extensive arrangement of this kind inay be placed between three thin glass rods, covered with sealing-wax, and secured in a triangle, by being cemented at each end into three equi-distant holes in a round piece of wood; or the plates may be introduced into a glass tube previously well dried, and having its ends covered with sealing-wax, and capped with brass; one of the brass caps may be cemented on before the plates are introduced into the tube, and the other afterwards; each cap should have a screw pass through its centre, which terminates in a hook outside. This screw serves to press the plates closer together, and to secure a perfect metallic contact with the extremities of the column. To fill the tube with discs, it is necessary to employ a cylindrical rammer of baked wood with flat ends, and when a small number (as about half a dozen discs) are introduced, they should be thrust down, taking care to ensure their perfect contact; and the operation of the apparatus will be ensured.

124. Soon after the invention of the column, Mr. B. M. Forster discovered that, when a sufficiently extensive series was put together, its electric power was sufficient to produce a sort of chime by the motion of a small brass ball between two bells, insulated, and connected with the opposite extremities of the column. He constructed a series of 1500 groups, and by its agency kept a little bell-ringing apparatus in constant activity for a considerable length of time.

125. Mr. Singer contrived an arrangement which is well calculated to form a perpetual motion, by excluding, to a very considerable extent, the operation of extraneous causes of interruption, and it at the same time renders the disposition of the apparatus rather elegant. A series of from 1200 to 1600 groups are arranged in two columns of equal length, which are separately insulated in a vertical position by glass pillars constructed on his principle of insulation; the positive end of one column is placed lowest, and the negative end of the other; and, their upper extremities being connected by a wire, they may be considered as one continuous colunin. A small bell is situated between each extremity of the column, and its insulating support, and a brass ball is suspended by a thin thread of raw silk, so as to hang midway between the bells, and at a very small distance from each of them. For this purpose the bells are connected, during the adjustment of the pendulum, by a wire, that their attraction may not interfere with it; and, when this wire is removed, the motion of the pendu

lumn commences. The whole apparatus is placed upon a circular mahogany base, in which a groove is turned to receive the lower edge of a glass shade with which the whole is covered.

126. If a column of about 1000 series is placed horizontally, with each of its extremities resting on a gold leaf electrometer, the electrometers will each diverge; that connected with the zinc extremity of the column will be positive, that connected with the silver extremity will be negative. If the column be very powerful, the gold leaves of the electrometer will alternately strike the sides of the glass, but this motion is soon stopped by their adhering to it. If the simple divergence only is produced, on touching either extremity of the column, the electrometer connected with it closes, and that at the opposite extremity has its divergence increased. This is analogous to the effect of the Voltaic battery when disposed in a similar manner; but the motion in the column is slower, which arises from the more imperfect conductors of which it is composed.

127. There is some cause, not yet perfectly developed, that appears to influence the power of the column to produce the motion of light metallic pendula. În the bell-ringing apparatus, for instance, though the motion always continues, it is much more rapid at one period than another, and the oscillations of the pendulum, though usually as uniform as that produced by mechanism, is on some occasions singularly wild and irregular. The frequency with which the gold leaves of an electrometer strike the sides of the glass, when connected with an electric column, is also different at different times; the variations observed in some experiments of M. De Luc are much more considerable than we have yet noticed, with the more powerful columns of Mr. Singer's construction.

128. De Luc proposed, as an interesting object of enquiry, to make regular observations on the action of the column, and the number of oscillations it would produce in a given time, at each observation. For this purpose a single column of from 1000 to 2000 series may be supported vertically on an insulating pillar. A bent wire, with a ball at its lower end, is to be connected with the upper extremity of the colurin, so as to hang parallel with, and at some distance from it; the ball at its lower extremity being diametrically opposite to a similar ball that is screwed into the lower cap of the column. To the same cap there is also screwed a brass fork with a fine silver wire stretched between its extremities; this is placed above the ball and projects farther from the column, so that when the pendulum moves towards the ball it strikes this wire first, and receives a kind of jerk, which prevents it from sticking. The pendulum consists of a gilt pith-ball suspended by a very fine silver wire, which hangs parallel to the bent brass wire, to which it is fastened at top; the arrangement is such, that the gilt pith-ball would be always in contact with the brass ball that proceeds from the upper extremity of the column, if the apparatus had no electrical power; it therefore always returns to this situation, when, after being attracted to the lower extremity of the

column, it discharges its electricity by striking against the cross silver wire.

129. There appears every reason to believe, that the action of a well-constructed column would be permanent. There is, however, a precaution necessary to their constant and immediate action; the two ends of a column should never be connected by a conducting substance for any length of time; for if, after such continued communication, it be applied to an electrometer, it will scarcely affect it for some time. It is, therefore, necessary, when a column is laid by, that it be placed upon two sticks of sealingwax so as to keep its brass caps at the distance of about half an inch from the table, or other conducting surface on which it is laid. And if a column, which appears to have lost its action by laying by, be insulated in this way for a few days, it will usually recover its full power. 130. There is another cause of deterioration which is more fatal; it is the presence of too much moisture. If the paper be perfectly dry it is a non-conductor, and will not therefore produce any action in the column; but this perfect dryness can only be obtained by exposing the paper to a heat nearly sufficient to scorch it, and in its dryest natural state the paper will be found sufficiently a conductor, even when, by exposing the paper discs to the heat of the sun, they have been so dried as to warp considerably. When the paper is sufficiently dry, the action of the column continues without diminution; and on taking such an apparatus to pieces after it had been constructed thirty months, no trace of oxidation was evident on the zinc plates.

131. The size of the plates in the column need not be large; Mr. Singer has constructed them of various sizes, and finds no proportionate advantage by extending the diameter beyond fiveeighths of an inch; they may even be constructed much smaller, and yet found to act with the greatest precision.

132. By connecting the extremities of a column of at least 1000 series, with the opposite coatings of a Leyden jar, during a period of from one to five minutes, a charge is usually communicated to it capable of affording a small but distinct spark, when the discharge is made by a wire that is not very thick.

133. Mr. Singer observes that the most extensive series he had ever made experiments with, consisted of 20,000 groups of silver, zinc, and double discs of writing paper. Its power was considerable. Pith-ball electrometers, with balls of one-fifth of an inch diameter, and threads of four inches long, diverged to the distance of two inches and upwards, when connected with its opposite extremities. An electrometer in the centre was not affected. When either extremity of the column was connected with the ground, the electrometer attached to that extremity closed, and the central electrometer opened with the same electricity, whilst that connected with the opposite extremity had its original divergence considerably increased; but the electro-motion was so slow, that some minutes were required to produce the full effect.

134. By connecting one extremity of the series with a fine iron wire, and bringing the end of

this near the other extremity, a slight layer of varnish being interposed, a series of minute bright sparks were obtained by drawing the point of the iron wire lightly over the varnished sur

face.

135. A jar containing fifty square inches of coated surface was charged by ten minutes contact with the column, so as to convey a disagreeable shock, felt distinctly in the elbows and shoulders, and by some individuals across the breast.

136. The charge from this jar could perforate thick drawing paper, but not a card. It had just power to fuse one inch of platina wire, of the th of an inch diameter.

137. Notwithstanding the considerable electric power of this combination, it had not the slightest chemical action; neither the best nor worst conducting media were affected. Saline compounds, tinged with the most delicate vegetable colors, were exposed under the most favorable circumstances to its action, and in some instances for many days, but no chemical effect was produced.

138. It therefore appears indispensably necessary to the chemical power of the Voltaic apparatus, that a liquid be interposed between each pair of its plates, whilst, for the pure electrical effects, the only condition appears to be the association of the two metals, and the connexion of the different pairs, by some conductor that does not interfere with their electro-motive power.

138*. The first experiments made upon the most pile in this country appear to have been performed by Messrs. Nicholson and Carlisle. After observing the effects then ascribed to the piles on bringing the wires from each end of the column in contact with a drop of water, they observed a disengagement of bubbles of some elastic fluid.

139. On closer examination they found the gas to be hydrogen. They then took a glass tube, about half an inch in diameter, into each end of which a cork was inserted, the tube being filled with water. Through each cork was introduced a brass wire, so that the ends of the wires in the glass were about an inch and threequarters of an inch. The pile employed consisted of thirty-six half-crowns, and as many similar pieces of zinc, and wet pasteboard. The zine end of the pile was then connected with one of the wires in the tube, and the silver end with the other, so that the circuit formed by the pile was separated by the water in the tube placed between them. A stream of bubbles was observed at the end of the wire, in the tube connected with the silver end of the pile. No gas was disengaged from the opposite wire, but it speedily became tarnished, first of an orange color, and ultimately black. The tube was then reversed, when it was observed that the wire, which in the first experiment became tarnished, gave out bubbles, while that which had before given out gas, in its turn became tarnished.

140. The emission of gas from the wire connected with the silver end of the pile was constant and uniform, except when a metallic circuit was formed between the ends of the pile, during VOL. VIII.

which no gas whatever appeared. It was observed that, when this metallic conductor was removed, the appearance of the gas was not immediate, since there was an interval of about two seconds between the removal of the wire and the appearance of bubbles. After the process had continued two hours and a half, a bulk of gas was produced equal to two-thirds of a cubic inch. This gas was mixed with an equal bulk of common air, and exploded on the application of a lighted taper.

141. These ingenious experimenters, supposing the phenomena in question to arise from the decomposition of the water, thought it surprising that the hydrogen should make its appearance at a distance of an inch and three-quarters from the point where the oxygen was disposed of.

142. They then made the same experiment with a tube thirty-six inches in length, but no gas was observed. When they introduced an infusion of litmus instead of pure water, they observed that the fluid in the vicinity of the wire connected with the zinc end of the pile became red, and hence were led to suppose that an acid had been produced. The fluid at the other wire was not changed, but gas, as usual, was evolved.

143. It may be proper to state that, in every apparatus constructed for practical purposes, there is a combination of three different substances in contact with each other, in successive groups; in general it is an arrangement of copper, zinc, and some conducting fluid. It is demonstrable that the primary source of the electrical power of the apparatus is the association of the two metals; and, according to Volta, the interposed fluid serves only as a conductor of the effect of one pair of metals to another. As far as electricity is concerned, this opinion appears to be correct, for the electrometer is acted on, whatever be the nature of the interposed fluid, and the degree of divergence is proportioned to the number of the plates. The electrometrical effects prove also, that, the arrangement of a series of zinc and copper plates, with an interposed fluid, forms a conducting column, which, in its insulated state, is positive at one extremity, negative at the other, and neutral in the middle. This may be easily shown by three gold-leaf electrometers, connected at the same time with an apparatus of 300 or 400 pairs of plates. The electrometer, connected with the copper extremity, will diverge with negative electricity; that connected with the zinc end will separate to the same distance positively; while that connected with the central plate of the series will not be affected. But, if either extremity of the battery be connected with the ground by means of a wire, the leaves of the electrometer connected with it will close; and those of the central electrometer will open with the same electricity, and to the same extent, whilst those of the opposite extremity will have their original divergence increased.

144. Hence it appears that there is a real electro-motive property in the apparatus, by which the zinc end constantly tends to become positive, and the copper end negative; and it is also obvious, that the extent of this operation,

H

at either extremity, is increased by connecting the opposite end with the ground. This last experiment, by which the central plate may be rendered either positive, negative, or neutral, at pleasure, proves also that the interposed fluid never acts as an insulator, for if it did so these changes could not possibly occur.

145. As the contact of either surface of the battery with the ground increases the electrical state of the opposite extremity, the same circumstance may be presumed to take place with every pair of associated metals, when their surfaces are in contact with a conducting fluid. Whilst the apparatus is insulated, the first zinc plate can only act on the electricity of its associate, the first copper plate; but the second zinc plate, through the conducting interposed fluid, can act on both these, besides its companion, the copper, and may therefore become more highly positive; and it is easy to conceive that such a repetition of action would be attended with an increase of effect, proportioned to the number of plates; and that the electrical tension of either end must be increased by connecting the other with the ground.

146. To ascertain if this principle really operated with a single combination, Mr. Singer took a pair of circular plates six inches diameter, very clean and smooth, one being formed of zine and the other of copper, and each provided with an insulating handle. When both plates were held by their insulating handles, and the zinc was successively applied to the flat surface of the copper, and after each contact made to touch the insulated plate of a condenser of six inches diameter; twenty contacts were required to communicate such a charge to the condenser as would occasion the leaves of a very delicate electrometer to separate to a quarter of an inch. But when the copper plate, instead of being held by its insulating handle, was simply laid on the hand, or on any similar conducting body, ten successive contacts of the insulated zinc plate, communicated a charge to the condenser, which occasioned the gold leaves to separate to the distance of more than half an inch. On repeating these experiments, with the variation of touching the condenser with the copper plate, held by its insulating handle, and brought in contact with the zinc plate, first insulated, and then uninsulated, similar results were obtained, but with the contrary electrical state. Hence the similarity of action in a single pair of metals, and a combined series, is sufficiently proved; and the preceding statement of the manner in which the electrical power is supposed to increase with the number of associated plates, is rendered highly probable.

147. So far the phenomena are sufficiently simple and consistent, for those described are not materially influenced by the nature of the interposed fluid, nor do they occur, but when the extremities of the apparatus are unconnected with each other, and consequently capable of maintaining the opposite electrical states. But the chemical effects, the shock, and the power of ignition, take place only when the extremities of the apparatus are connected by some conductor, and are also materially influenced by

the nature of the interposed fluid. If these effects then, are produced by electricity, they can only result from its circulation in the apparatus; and, as there is no reason to suppose that the electro-motive power of the associated metals ceases when there is a conducting communication between their opposite surfaces, but rather that it is accelerated by such a circumstance; that very acceleration may be the cause of the phenomena, and the effects observed correspond very nearly with such an idea; for, if it be admitted that the connexion of the opposite ends of the Voltaic battery by a conductor, occasions a current of electricity from the positive to the negative, that current must be more rapid, in proportion as the conductor is more perfect. Now it is found that the chemical effects are most considerable, and more promptly produced in fluids of the highest conducting power; thus the quantity of gas liberated in a given time from common water, is greater than from distilled water; saline fluids furnish more than common water; solutions of alkali more than saline fluids, and acids more than alkalies: and, as the effects of a simple combination are influenced by the same causes as those that operate with a series, the fluids that are susceptible of the most rapid decomposition are also most active in exciting the chemical effects of the battery, when employed as the connecting medium between its plates.

148. Acids are of all other fluid bodies, excepting metals, the most perfect conductors, and the chemical effect of the battery is more powerfully excited by them than by any other substances; it is possible that their chemical action on the zinc may have some share in modifying the quantity of electricity, or the rapidity of its motion; but it is certain that the effects are not in proportion to the chemical action; sulphuric acid, for instance, acts as powerfully on the zinc as nitric or muriatic acid, but it is not so active in producing the chemical agency of the battery: in like manner the alkalies, which exert a very trifling action on the battery, excite its powers with greater energy than many saline fluids which are more efficient as chemical agents.

149. The ignition of wire, and of charcoal in the Voltaic circuit, is conformable to this view; these substances are the most perfect conductors known, and, when made the medium of communication between the opposite ends of a battery, must accelerate its electro-motive power to the greatest extent. The rapid circulation of electricity, thus obtained, produces ignition, if the conductor be not too large in proportion to the quantity of electricity; but, within this limit, the effect will be greatest with the thickest wire, because the acceleration will be more considerable in proportion to the facility of transmission. There is, perhaps, no other view on which the continued ignition of wire, and the increased action of large plates is so intelligible.

150. The cessation of chemical agency, and igniting power, as the chemical action of the acids or other menstrua declines, may arise from the total change which then occurs in the nature of those fluids; their conducting power is much diminished, and they may possibly, by the change

in their chemical properties, acquire some faculty of electro-motion subversive of the effect of the combined metals.

151. The extensive experiments of Messrs. Hisinger and Berzelius, confirmed by the researches of Sir H. Davy, had demonstrated the constant separation of oxygen, and compounds in which it prevailed, at the wire proceeding from the zinc surface, and of hydrogen and other inflammable matter, at that connected with the copper surface; and, at this latter, alkali was also frequently found, and, from analogy, it was in consequence concluded, that the alkalies probably contained a considerable proportion of some inflammable substance.

152. This conjecture was confirmed by Sir H. Davy in 1807: he found that a thin piece of potassa or soda, slightly moistened by exposure to the air, and placed between two conductors of platina, proceeding from the opposite sides of a powerful Voltaic apparatus, was resolved into a peculiar metallic substance highly inflammable, which appeared at the negative surface; and oxygen gas, which was evolved at the positive surface. By an extensive series of experiments, it was shown that these bodies are, in reality, metallic oxides, and that the proportion of their constituent parts is somewhat different, being in round numbers, for potassa six parts of metallic base to one part of oxygen, nearly; or it may be stated that potassa is composed of eighty-six parts of metal, and fourteen of oxygen in each 100 parts. The proportions in soda are nearly seven parts metal to two of oxygen; or seventy-eight metal and twenty-two oxygen, in each 100.

153. The metal obtained from potassa is called potassium; it is lighter than water in the proportion of eight to ten. At common temperatures it is solid, but soft and plastic. At a temperature of 150° it becomes fluid, and evaporates at a heat rather below redness. In color it nearly resembles silver, but it tarnishes immediately when exposed in the open air, and can only be preserved under naphtha. Its attraction for oxygen is so powerful, that it will detach that substance from almost all its combinations; and the result of this action is its consequent oxidation and reconversion into potassa. If thrown into water it immediately inflames, floats upon the surface, and burns with a mixed flame of white, red, and violet; rendering the water, in which the experiment is made, alkaline. Similar phenomena ensue when it is brought in contact with ice. When moderately heated in oxygen gas it inflames and reproduces potassa. Its action on water is always attended by the decomposition of that fluid; hydrogen is evolved, and the oxygen combines with the potassium to form potassa. By measuring the quantity of hydrogen separated from water, by the action of a given weight of potassium, the quantity of oxygen that metal combines with to form potassa may be readily learnt. Each grain of potassium detaches about 106 cubic inch of hydrogen gas, and consequently combines with half that quantity of oxygen.

154. The metal obtained from soda is named sodium; it is rather lighter than water, nearly as 09348 to 1000. It has the color of silver; is less fusible than potassium, but tarnishes in air

in the same way. It is fluid at the temperature of 200°, and passes into vapor at a strong red heat. At common temperatures it is a soft metal, and a globule of it may be easily spread into a thin leaf by the action of a knife. It decomposes water violently, and floats on its surface, but does not inflame; the water is rendered alkaline, and, when examined, is found to contain pure soda. It acts nearly in the same manner as potassium, but with less energy, on most substances, and must consequently be preserved under naphtha. When thrown on the surface of nitric acid it inflames, and burns with great brilliance; it also occasionally scintillates when thrown upon hot water. The proportion of oxygen with which it combines to form soda, may be learnt by noting the quantity of hydrogen evolved from water by a given weight of the sodium.

155. Both these new metallic substances unite with mercury in various proportions, and form amalgams which decompose water, but more slowly than the metals themselves; these amalgams act upon all other metals, even platina and mercury.

156. The decomposition of the alkalies may, by care and attention, be effected with a battery of fifty pairs of plates of three or four inches square; but the results are rather uncertain: 200 plates form a very efficient arrangement; they should be excited by a weak acid mixture (about one part strong muriatic or nitrous acid, to thirty parts of water). A plate of silver or platina being connected with the negative side of the battery, a thin piece of pure potassa or soda is to be placed upon it, and a platina or silver conductor, proceeding from the positive side of the battery, is to be brought in contact with the upper surface of the alkali, which soon fuses at the points of contact: metallic globules shortly appear near the negative surface, and gradually increase in size, until a crust of alkali begins to form on their surface; at this moment they should be removed by the point of a knife, and instantly plunged under naphtha; or, if the experiment be merely intended to demonstrate their production, they may be brought in con

tact with the surface of water or nitric acid. It

sometimes happens that no globules appear; but if the contact be preserved for some time, and the alkali be afterwards raised, several will be found imbedded in its under surface. If the action of the battery be strong, it also sometimes happens that the globules inflame, and even detonate at the moment of their production; it is therefore advisable not to bring the eyes too near during the experiment, or else to cover them with glasses. These experiments always require great care to insure their success, which a trifling variation in the power of the battery, purity of the potassa, or moisture of the atmosphere, may prevent. -Soda is rather more difficult to decompose than potassa, and therefore requires to be employed in thinner pieces; the pieces of potassa should rarely exceed a quarter of an inch in thickness, and those of soda oneeighth of an inch.

157. To prevent the loss of the alkaline bases during their separation, by the powerful action of the air upon them, it has been proposed to effect