This may he confirmed by means of the appara tus represented in Plate CX XIII. Fig. 11, consist ing of two parts, viz. a movable conductor, A C B, and a multiplying wire D E F G. The movable wire is terminated by two steel points at A and B, which are to be placed in two small steel cups filled with mercury, and communicating with a galvanic apparatus. The multiplying wire is preferred to a straight one, in order to increase the effect. The upper part D E, of the multiplying wire is placed at the same height as the branch B C of the mov able conductor; but in such a position that both conductors prolonged would form an angle. The extremities and G of the multiplying wire are to be put in communication with a galvanic appa ratus.
Mr. Ampere, to whom we are indebted for the discovery of the mutual attractions and repulsions of the electrical currents, considers the law of this action as a fundamental one, at least so far as our present knowledge extends. He thus admits no rotative action in the electrical current, but he transports it to the magnet, in which he supposes electrical currents, revolving in planes perpendicu lar or nearly perpendicular to the axis of the mag net. At first he supposed that all the currents had their centres in the axis, and were situated in planes perpendicular to this axis, but as he soon found that this would not represent the phenomena, he supposed that each atom of the magnet was sur rounded by electrical currents, still revolving in planes perpendicular to the axis of the magnet. When Mr. Poisson, however, showed, that in con sequence of this view the greatest effect of a mag netical bar would be placed in its extremity, con trary to experiment, he changed this supposition, and at present he is of opinion that the currents are situated in a plane somewhat inclined to the axis of the magnet.
By these suppositions, and a considerable exer tion of mathematical skill, he is enabled to make this view represent well enough the phenomena, though his theory is very complicated. It is not necessary here to enter into a discussion on all the points of this theoiy, as simple consideration of the fact upon which it is founded will be sufficient to decide the question.
Let us suppose that electromagnetism had not been discovered before the discovery of the mutual action of electrical currents, the application of the common philosophical rules should enable us to discover therein the rotative character of the ac tion. The fact is, as above mentioned, that parallel currents attract each other when they have the same direction, and repel each other when they have op posite directions. Now it is to be remarked, that two parallel things of the same direction have their opposite sides placed against each other: the left of the one is nearest the right of the other: but two parallel things of contrary directions have their si milar sides turned against each other: right against right, or left against left. Thus the fact reduced to the simplest philosophical expression is that two points of electrical currents repel each other by their similar sides, and attract each other by their opposite sides. The most direct enunciation of the experi mental result cannot here be considered as at the same time the expression of the philosophical one; for it is evident that two parallel things cannot aet upon each other immediately, but only by some transverse action, which here shows itself as con sisting of attractions and repulsions in opposite di rections, or in other terms, as having polarity. But
such contrary powers forming a circle, should keep themselves in equilibrium, and produce no effect without their limits, were they not in motion. Thus the very experiment of Mr. Ampre should, in the absence of all other evidence, be sufficient to prove, that the electric current contains a revolving action, exhibiting every appearance of polarity. We do not mean to ascertain the nature of these attractions and repulsions; but it has been our object only to point out the more immediate consequences of the facts.
Electromagnetieal Currents produced by heat.
Dr. Seebeck, in his researches upon electromag netism, extended at the same time his investiga tions to the laws of galvanic action, and among these to the influence of heat in galvanic arrange ments. Some phenomena here occurred to him which led him to think that two metals, forming a circuit, might produce magnetism when the equi librium of heat in it was disturbed. Experiment confirmed this opinion. Plate DXXIII. Fig. 12. re presents such a circuit; let ABC be a piece of bis muth, and ADC a piece of copper, and let one of the junctions, .11 for instance, be heated, an electri cal current will be established, which here can only betray its existence by the magnetical needle; this indicates all the magnetical properties of an elec trical current, and, in the instance here mentioned, the current goes into the heated junction from the bismuth to the copper. Dr. Seebeck is not inclined to consider the effect thus produced as a true elec trical current, but an effect sui generis: and indeed we have not hitherto been able to discover in this circuit either any chemical effect, nor heat or light; still we can represent all the phenomena of Dr. Seebeck's circuit by the same terms as those of the common electrical current: and in the explanation of all the facts, it will appear highly probable that this current is truly a particular kind of electrical one. Professor Oersted has proposed to call the current discovered by Dr. Seebeck the trical current, and in consequence of this to distin guish the action hitherto called Galvanism by the name of the current. Hence we have now the names and city, to which we could add the name city for the electricity produced by friction. Dr. Seebeck has made a very considerable number of experiments upon the thermoelectricity produced by the metals and other perfect conductors. In a circuit containing bismuth, together with one of the other metals, he finds that, in the heated junction, the current goes always from the bismuth to the other metal; of course the bismuth loses, at that point, positive electricity. This we shall, for short ness sake, express thus: bismuth becomes negative with all other metals in the thermoelectrical cir cuit. In the same sense tellurium may be said to become positive with all other metals. It appears already by these two examples, that the thermo electrical order of the metals is not the same as the hydro-electrical; and indeed the experiments of Dr. Seebeck have proved that these two orders are dis• crepant throughout.