foot of surface, and immersed it in a mixture of one part acid in five of water, and found that one pound of zinc was dissolved in fifteen minutes; this then was the measure of the maximum amount of electricity a battery of such size could generate in fifteen minutes; but the whole amount a practical construction of plates of such size could generate, would not be the tenth of that amount, and it is easily foreseen that whatever the amount might be, a cable of such length and favorable arrangements for "static induction" could be constructed that all the electricity which could be generated by a series of such pairs, in fifteen minutes, with least conduction resistance, could be disposed on its sides before it would constitute a charge, and consequently the charging of such cable with that battery would require more than fifteen minutes.

The whole quantity of electricity which any battery will generate in a given time under any given circumstances will be proportional to the size of the battery plates and the interior conduction resistance, and consequently the time required for any battery to charge the Atlantic cable will be in an inverse relation to the size of the plates-the number of pairs not affecting this: it merely determining the tension to which the charge rises; though on this latter depends the available quantity at the receiving end for recording.

A battery to charge the cable in the least time should have the plates of unlimited size: this would not be possible, neither would it be practical to have them bear any small proportion to the four acres to be charged. The minimum time however may be nearly obtained by using a sufficient reservoir to collect the electricity evolved from convenient sized plates and thus working the cable by means of an immense Leyden battery. For this purpose I would construct a Leyden jar of metallic sheets and varnished paper or silk, of several acres in extent, and connect the alternate metallic sheets with suitable bars for attaching the battery with the earth and cable. It will doubtless at once be perceived that it would take quite a time to charge such a Leyden jar-say eight acres-by the battery, and here it will be seen that the mask is up which has disguised the time of charge (the socalled wave time). To merely enlarge the battery plates somewhat above their present dimensions would doubtless obviate the difficulty to a proportional extent, but full enlargement should be made, for it will be desirable to work as rapidly as possible. With such a Leyden battery as I propose the tension which propels the electricity into the wire could at no time fall more than one third, and its effective power in charging the cable would be some thousand times greater than an arrangement of an intensity series of small pairs. In short, it is apparent that by such an immense Leyden arrangement and a sufficient battery

to charge it, the electricity would find its way through the conductor or through the insulation.

From the tone of the European publications I have been led to infer that they do not consider it possible to work the cable in a prompt and efficient manner by means of the battery. The slow and imperfect action of the Balaklava and some other cables seems to have early thrown a damper on the prospects for submarine lines, and the suspicion seems to have become conviction after the publication by Mr. Faraday of his experiments, together with those of M. Melloni and Mr. Clark; according to those experiments it is demonstrated that increasing the number of pairs in the battery, or increasing the "intensity," as this operation is often improperly called, does not diminish the "wave time" or the time for the electricity to flow through the conductor. By the experiments on the subterraneous wires of the Manchester telegraph, the time for the current to manifest itself through 762 miles of wire, was the same whether 30 pairs, 50 pairs, or any number of pairs up to 500 were used. If I do not mistake the tone of these, and all the subsequent papers on the wave time for submerged conductors, these experiments are regarded as decisive against the possibility of shortening the time by any modifications of the battery. At the results of those experiments M. Melloni appears to express astonishment, and considers it as opposed to the laws of Coulomb and others, and against the received ideas of quantity and intensity. Mr. Faraday appears to be disappointed however, he having predicted if I rightly apprehend his meaning, that the current from the greater number of pairs should have manifested itself more quickly than that from the lesser, and this nonconformity of the experiment with his predictions he attributes to the fact that while the intensity of the current was increased, its quantity was also increased, and considers with the increase in the number of the pairs, their size should have been diminished in order that only the same quantity of electricity in a more intense condition might have been permitted to traverse the conductor. From all this it is evident that those views offer but small prospect of the battery ever serving to work the submarine telegraphis. I cannot however, but express astonishment that any other effect on the wave time, from increasing the number of pairs, than that exhibited in the experiments of Mr. Clark should have been expected, considering that at the first moment of the contact of the battery with the insulated submerged conductor, the insulation in undergoing polarization is acting as a conductor; thus at the beginning of contact, the wire acts not only as a conductor of the capacity of its cross section, but also as though it had a solid. section of the area of its whole surface, thus acting on the battery, as a short thick wire (in the case of the Atlantic cable

as a wire having a solid section of 185,000 sq. ft.) in which the resistance exterior to the battery may be considered as nothing, and the whole resistance to the generation of the electricity as being only that which lies within the battery: now taking Ohm's formula for the electric current in which E represents the R+r

E

electromotive force of the chemical affinity, R the resistance to conduction due to the materials of the battery, and r the resistance of the conductor which completes the circuit by connecting the poles of the battery, and considering the effect of multiplying the number of the elements we have and considering r as

nE E

nE

nR+r'

being indefinitely decreased by diminution of the length of the conductor, or by increase in its solid section, which are the conditions that obtain while the cable is receiving its static charge, we may consider it as =0, and remove it from the equation, and then we have = in which it is evident a train of batteries nR R in the conditions which obtain at the first moments of contact with the cable, can generate no more electricity than a single cell attached to the same conductor. It is here evident that the facts arrived at by Mr. Clark, so far from being a cause of wonderment, are but what the circumstances should have suggested, neither do they call for a modification of Coulomb's laws of induction, or a change of our ideas of quantity and intensity, as suggested by M. Melloni; much less do they indicate that a diminution of the size of the battery plates would hasten the current as suggested by Mr. Faraday. When a battery has the circuit open we conceive that the terminal plates have a tension proportional to the number of elements in the train, T=n E; and as the whole quantity of electricity they contain will be the product of their surface by the tension, Q=TS, it is evident that when the battery touches any inductive surface of great extent, such as that of the insulation of the cable, will be divided with it, and as S compared to that surface is insignificant, and as S is constant, Twill fall enormously, as has been before shown for another purpose; now when the terminals of a compound battery, are connected with a stout wire, it is known that the tension falls to that of a single pair of plates, it is therefore evident that contact of the battery, with a large inductric surface, is equivalent to connecting with a non-resisting conductor. In this state of the battery it is evident that the electricity can be conveyed into the conductor only as fast as the battery can admit of the diffusion within, or bring the electricity to the terminals, in which case, as has before been shown, many pairs can generate no more electricity than a single pair, and consequently the tension will be below the maximum of a single pair,

in proportion to the smallness of the plates. Is it any wonder that with the feeble tension produced under such circumstances, such a long time is required for the battery to charge the cable?

For the sake of illustration I have considered r=0; such however is not strictly the case, the wire offers some resistance and the gutta percha offers some resistance to polarization, the tension of the terminal plates will therefore rise in proportion to this resistance until ultimately it reaches the maximum tension due to the number of electromotive elements in the train. If we consider furthermore that with any finite resistance the tension will rise in proportion to the decrease in the value of R, or that the extreme tensions will be inversely proportional to the internal conduction resistance of the battery-or if in the formula Q we give to r a small value and diminish R by enlarging the size of the battery plates, so as ultimately to obtain

nE

nE

nR+r

nR+rm Q-we see plainly that by enlarging the number of

m

electromotive elements in the battery, and at the same time increasing their size, any required effect can be produced on the cable.

From all the above it is evident that the time for any battery to charge a Leyden jar will be in a certain inverse proportion to the size of the battery plates; the charging of the cable is similar to the charging of the jar, excepting that it is complicated by the resistance of the wire, and the time will be a function of the internal conduction resistance of the source of the electricity. The general mathematical considerations must apply to every electrical action, whether it be produced by thermo-electricity, by chemical action, by magneto-electricity, by induction coils, or by the lightnings from the clouds. But as the battery is the only effective source of electricity, and as it is the cheapest also, reason calls for the construction of a sufficient battery, for working the cable in preference to all secondary contrivances.

No one at all acquainted with the subject will doubt that such an immense Leyden jar as I proposed above for passing our longitude signals, if charged to a high intensity, would more than all other things be fitted for affecting the cable. If the great jar were not discharged in charging the cable, it would continue steadily to work it in the transmission of messages; a jar that would remain thus continually charged would be a generator of electricity; such is a galvanic battery, and by increasing the number and size of the plates, the battery is made into such a continuously charged jar capable of producing any extent of effect.

I have not considered the fact that by the resistance of the wire of the cable, and the reaction of the parts which first receive the static charge, the flow of the electricity is broken up into waves, and that consequently the whole of the insulation of the cable does not perform its full inductive office at once. I have necessarily been confined to general principles and admitted conditions of electrical action, the deductions however are the same as though the whole inductive surface of the cable were simultaneously charged, for the degree of charge given to the first portion determines the rapidity with which the successive portions receive their charge.

The plan I propose for large plates or an electrical reservoir has advantages also for the discharge. Already it has been proposed and is practised I believe to discharge the conductor by making contact with the opposite end of the battery. As the plan I propose offers such greatly increased electrical conditions, it will correspondingly decrease the time of discharge.

I have no doubt that you will at once coincide with me in the correctness of the views of the cause of the slow working of the cable. I think you have previously perceived that a great fall in the tension must take place at the moment of contact, and the consequent importance of having large terminal surfaces to hold a sufficient quantity of electricity to maintain the tension. The expression by you of such views I take it is what Dr. Gould refers to in his report on the telegraphic operations for longitude.

I have already extended this letter to a greater length than consideration for your valuable time would justify; but that I may succeed in presenting the correctness of my views on this important matter of the Atlantic telegraph let me trespass on you for a minute longer, to show that these new views are in fact old, that all electricians acknowledge them in their writings, and that the phenomenon of the delay of the current is amongst the earliest experiments in electricity.

When the knob of a DeLuc's column is brought into contact with a gold leaf electroscope, the leaves diverge. If while the leaves are diverged the terminals of the pile are brought in contact, one with the inner and one with the outer coating of a Leyden jar, the leaves collapse, and the jar is found feebly charged; if the pile is now detached from the jar the leaves will begin gradually to diverge again; if however the pile is left in connection with the jar the leaves will after some time (several minutes) diverge again. Now the time between the collapse of the leaves and their regaining their divergence, is the time required for charging the jar, to the tension due the pile:-that is, it is the time required for that generator of electricity (the DeLuc's pile) to generate the quantity of electricity required to charge that jar to the tension due to the pile. Not a single

SECOND SERIES, VOL. XXVII, No. 80.-MARCH, 1859.