it may appear somewhat strange when we say that many months elapsed before that gentleman could get a competent firm to carry out his patent. People will sometimes be slow in yielding assent to anything that clashes with their natural conservatism. Franklin himself was made the object of some harmless pleasantry on account of his ideas about lightning-rods by one of the most distinguished bodies in Europe, our own Royal Society. But M. Werdermann ultimately found a firm able and willing to undertake the construction of his machines. Public interest was soon awakened by the results obtained by Mr. Conrad W. Cooke, in presence of some of the leading scientific men of the day and the representatives of the press. We had the privilege of examining the apparatus at our leisure, and of witnessing on three consecutive evenings a series of experiments, of which we shall speak after we have briefly discussed the technical parts of the apparatus.

It is necessary to premise that the construction of the machine depends upon the uses to which it is to be applied. If required for electro-chemistry, it must give quantitative effects; and if for illumination, tensional. As a general rule, these may be attained by varying the length and gauge of the wire and the connection of the bobbins; but besides these, other changes are introduced into this machine, which accordingly modify its appearance. The one shown in perspective on the accompanying plate (Plate CI. fig. 1) belongs to the second class. It consists essentially of several bobbins or armatures rotating before the poles of two rows of cylindrical electro-magnets. These occupy a vertical position, with similar poles opposite, but not in contact with, each other. They are 3 ft. 75 inches high, and 35 inches in diameter. The height of the machine is 4 ft. 1 inches, and its weight one ton. The armatures form quite a novel feature in magneto-electricity, and the idea is entirely M. Gramme's. It occurred to him that Siemen's long cylinder might be replaced by a ring. Accordingly, he coiled insulated copper wire around this core, and imparted to the whole a movement of rotation in a plane perpendicular to that of the adjacent electro-magnets, when he obtained unintermittent currents with maximum effect. Let N and s (fig. 3, Plate CI.) represent the north and south poles of one of the magnets. They will induce consequent poles at ss and nn in the adjoining portions of the ring; b and c on the magnetic equator will be the neutral points. The movement of the coil before N will induce in the wires currents which are positive in the upper quadrant, Nb, and negative in the lower Nc. Owing to the opposite polarity of s, a + current flows through sb, and a through sc. As consecutive parts of the helices. are constantly coming within the influence of N and s, it

follows that the currents will flow on without interruption as long as the rotation continues. The familiar class experiments, of which we have already spoken, go far to explain most of the phenomena of magneto-electricity; it will not, then, be necessary to dwell further upon primary principles. However, as it may not appear obvious what cause is immediately concerned in the production of the Gramme currents, we shall mention an experiment made by M. Gaugain, which will give a clear view of the subject. That distinguished physicist procured a ring of soft iron, and wound it with silk-covered copper wire in such a manner that the helix and its core might rotate either separately or simultaneously. This done, it was observed, 1st, that the strongest current was developed when the helix alone was displaced; 2ndly, that the current was weaker when the helix and its core rotated together; and 3rdly, that a very feeble current, inverse of the other two, was produced when the ring alone revolved. From this we gather that the currents owe their origin less to the displacement of the consequent poles than to the rotation of the helix. It would even appear as though the changes constantly taking place in the magnetic condition of the ring oppose the development of the currents.*

The wire which covers the ring is not one continuous piece, but is wound in lengths of about 10 yards each. The ends of two contiguous lengths are brought out from the bobbin and joined to metallic sectors, which are connected to as many copper conductors, placed axially on the spindle of the machine. These are insulated from each other by layers of silk, and it is essential that they be sufficiently numerous to form a compact cylinder, and that the insulating material be very thin. The currents are collected at the neutral points on the axis by metallic brushes, consisting of silvered copper wires held together by adjusting screws. These accessories ensure good soft contact; and when duly attended to, prevent the occurrence of sparks from the extra-current, which are always injurious to an apparatus. There are four of these brushes; two to collect the current, which is made to circulate round the iron cylinders, and the other two to collect the useful current.

The horizontal section (fig. 2) shows the number and disposition of the coils. When these are set in motion, very weak currents are generated by the remanent magnetism in the soft iron. The current from the bobbin near the drivingwheel is transmitted round the cylinders, and thus augments their magnetisation. This new increment now induces a stronger current in the bobbin, which in its turn excites a

Compare Dr. Ferguson, "Electricity," Appendix, p. 263.

greater intensity in the electro-magnets. This mutual action soon builds up a very powerful battery. The other two bobbins meanwhile produce the current which is used externally.

It is customary in other magneto-electric engines to excite the fixed electro-magnets once for all by means of a galvanic battery, the infinitesimal trace of residual magnetism sufficing to work the machine on subsequent occasions. Now a mere accident has led M. Gramme to dispense entirely with the use of the battery, for at the moment he was going to connect the electrodes of a few Daniell's cells with the machine—which had been previously set in motion by an attendant-he found it evolving a strong current. Terrestrial magnetism had here. anticipated him, and elicited in the soft iron the power he sought to induce.

One of the great drawbacks in the machines of Wilde and Ladd is the heating of the armatures caused by the rapidity of their rotation, which varies from 1,000 to 2,400 revolutions per minute. This involves certain inconveniences, which have militated against their extensive adoption for practical purposes. The normal speed of the Gramme is only 300 revolutions per minute; and it may be noticed that it is only when the external work is not proportional to the current generated that the temperature of the coils is slightly raised. By a proper management of the driving-power, but little if any heat may appear in the armatures, and a large fraction of the motive force may be converted into useful work. In no case, evidently, can the energy derived from a magneto-electric apparatus exceed the power expended in producing it.

There is another striking feature in this machine besides the continuity of its currents, viz., the possibility of increasing the number of electro-magnets without material inconvenience. The advantage that may be drawn from this is that several distinct currents may be derived from a single machine and applied to different purposes.

The current produced is equal to that of a Bunsen battery of 525 cells, arranged in five rows. Its tension is therefore that of 105 cells, and its quantity that of five. The wire on the coils is No. 12, and weighs 165 lbs.—that on the electromagnets is No. 11, and weighs 550 lbs.

A fair knowledge of the vast energy of this machine may be derived from the experiments to which we referred in a preceding paragraph. A platinum wire, No. 18 gauge and 15 ft. in length, was raised to a glowing heat, and the beautiful experiment illustrative of the specific resistance of platinum and silver was readily performed with 18 ft. of alternate pieces. Twenty-two and even 32 ft. of high conductivity copper wire, 96 per cent. purity, was stretched between the terminals-the middle

portion being wound in the form of a helix-and easily brought up to a white heat. A round file, inch in diameter and 4 inches long, was burnt up in about five minutes, and a small diamond was volatilised in a few seconds. We had brought with us a De La Rue discharger, containing cylindrical pieces of zinc, steel, and copper 475 in. in diameter and about 3 in. long. A dazzling light of various tints was instantaneously emitted as the different pairs of metals were brought into circuit, while large beads of the melting mass dropped down. M. Gramme records the following:

Heated to redness 40 ft. copper wire, 027 inches in diameter.

To give an idea of the efficiency of the machine for illuminating purposes, the electric light was projected into the York Road, when peculiarly dark and elongated shadows of vehicles and by-passers were cast on neighbouring houses. The street itself sparkled as though sown with a thousand oriental sapphires. The lamps grew dim, and shadows from the gas flames were in several instances cast on adjoining walls. The bobbins were making 300 revolutions per minute, and the brilliancy of the light was estimated nearly equal to 900 Carcel burners, or 8,640 sperm candles.

The image of the carbon points on the screen was highly interesting. One could plainly distinguish the white glow of the positive carbon, the frequent transport of incandescent particles generally in the direction of the current, and finally the sharp point of the negative carbon and the conical cavities of the positive-the whole lit up by that wonderful arch of flame, to whose excessively high temperature the most refractory metals must yield.

The principal spectra exhibited were those of platinum, silver, potassium, sodium, magnesium, and strontium. Some of the first authorities in spectrum analysis grew quite enthusiastic over the extraordinary brightness of several bands, and were able to detect many lines never before observed.

The Gramme machine has thus already contributed to extend the limits of science. What services it may yet render is a problem for the future. There are several physical questions that require to be investigated, and a few chemical controversies that need to be sifted and set at rest. Perhaps when the first excitement that is naturally caused by the introduction of a remarkable invention subsides, some attention may be paid to those subjects, and we should not be surprised to find the fallacy of certain hypotheses and the inaccuracy of certain numerical data clearly demonstrated. Nay more, we think that

Nature will not refuse to unfold some of her secrets to the new Davy who will best know how to use the great power placed at his command.

While on the one hand this apparatus affords new means of research, on the other it is itself a striking example of the conservation and correlation of forces, which it is the tendency of contemporary science to establish. The mechanical force that drives the coils is converted, under the eye of the observer, into a stream of electricity. Heat disappears in the boiler of the steam-engine, and Protean-like, reappears at the terminals as electricity, magnetism, light or heat. Here we have a strong argument for the dynamic theory, which would make four of the most important branches of general physics only different molecular modifications,-only different manifestations of the

same cause.

Viewed from a utilitarian standpoint, we should say that the comparatively low price of this machine-about £400*-is a close approximation to the production of cheap electricity. Now-adays, with our strikes and our unions, every one is more or less of a financier, and wants a full, often an ideal, equivalent for his outlay. This economical recommendation is then not one of the least cogent, and it only remains to be shown by actual work that this machine may be used with decided advantage in various departments of the arts. Its efficiency for a few of the most important electro-metallurgical purposes has been tested in the extensive and well-known works of M. Christofle, Paris. The tension of the apparatus used was equal to that of two ordinary Bunsen cells, and the quantity to thirty-two. The whole weighed 9 cwt., and a one horse-power sufficed to drive the coils at a rate of 300 revolutions per minute. At this speed, it was calculated to deposit at least 20 ounces of silver per hour. Two series of electro-plating experiments were made with a view to ascertaining the relative electrolysing effects of the Gramme and Wilde machines, and the result was decidedly in favour of the former. As this is a subject of interest to the general scientific reader, and of especial importance to the community at large, we shall give the details of the elaborate trials to which we refer.†

*This is the cost of the one which we have described. The demand will, of course, cause a great reduction in this figure.

"Comptes Rendus Hebdomadaires des Séances de l'Académie des Sciences." Dec. 2, 1872.