The experiments of Riess and of Henry were therefore nearly simultaneous, as were the subsequent announcements. The article mentioned anticipates, however, much that is discussed in the following sections of this report, founded on later publications of Riess and others. Thus experiments upon screening effects, upon secondary conductors at different distances, and upon the difference in magnetism, were recited. The latter of these, in connexion with the matter in § 70, throw additional light upon the apparently abnormal development of magnetism. But the whole set of experiments, and the deductions from them, were given as a sequel to similar investigations upon secondary currents with galvanic electricity; severed from this connexion much of their value would be lost, and to reproduce the whole, together with later researches in the same line, would take up more space than can be spared in the present volume G. C. S.] Riess proved the existence of the secondary current in the following manner, (Pog. Ann., XLVII, 55.)

Let A A, in fig. 59, be a 1 a copper wire wound spirally about a glass tube and introduced into the conducting circuit of a battery; A A consequently is the main wire. A wider glass tube is passed over the main wire, and upon it the sec ondary wire B B is wound, leaving its ends hanging free. The ends of a third spiral CD, also wound

upon a glass tube, are to be fastened at a and b.

The connection at b being severed, and the ends of the wire separated a little, a spark is seen to pass at b when the battery, with a sufficiently strong charge, is discharged through the main wire.

This spark is a proof of the existence of the secondary current. A passage of electricity from the main to the secondary wire cannot take place if the secondary spiral be kept at a sufficient distance from the ends of the glass tube on which it is wound.

A steel sewing needle placed in the glass tube of the spiral CD, which we will call the magnetizing spiral, will be magnetized by the secondary current.

An electrical air thermometer inserted in the secondary circuit indicates heat produced by the secondary current.

Figure 59 represents the form in which Riess first arranged his experiments. Afterwards (Pogg. Ann., L, 9) he gave the spiral a

more convenient form.

In a disk of wood, consisting of three pieces glued together, the diameter of which depends upon the size of the spiral to be formed, concentric grooves are to be cut and made into a spiral, by joining each circle with the following one by a curved groove; the innermost

Fig. 60.

circle is joined to the second by the groove c d, (figure 60,) the second to the third by e f, &c. In these grooves a copper wire about half a line thick is so laid as to make a spiral. One end of the wire passes through the disk at a, and along the under side to z. From a the wire coils out to c, from c to d, from d to e, f, &c.; x y is the other end of the wire thus wound in a flat spiral.

The disk is covered with a thin coat of pitch before placing the wire upon it.

The wire being fastened by the superposition of a hot metallic plate, the spaces between the rings of wire will

be filled up with the pitch; a heavy heated plate laid on the disk will make the spiral perfectly level. This spiral is now blacked with coal and pressed upon another wooden disk to

get the marks for a second spiral, which must correspond with the first as nearly as possible

The disks are now fastened to glass supports, their planes being vertical. They are arranged upon the same stand opposite each other, and so that they can be approached and separated at pleasure. This arrangement is represented in figure 61.

Another arrangement of the flat spiral, much more convenient for many purposes,

Fig. 61.

shown to me by Professor Eisenlohr, of Carlsruhe, is represented in figure 62. One of the spira's is fastened on an upright glass support in a horizontal position. The second

spiral is fastened in the same manner on a glass rod, which has no foot; it is placed over the other, like the upper, over the lower condenser plate.

The distance between the spirals can be changed by placing glass plates of different thicknesses between them. For greater distances pieces of varnished wood having any desired thickness are interposed.

The ends of the wires are provided with screw clamps z and y, by means of which the spiral can be connected as may be desired.

Fig. 62.

Placing y and z of the lower spiral from one to two lines apart, and separating the two spirals by a glass plate, a spark will be seen to pass between y and z on discharging a jar, sufficiently charged, through the upper spiral. The spark is produced by the secondary current.

§ 57. Magnetizing by the main current.-To avoid false conclusions in regard to magnetizing by the secondary current, magnetizing by the main current should first be properly investigated.

Such an investigation was first made by Savary. Riess repeated Savary's experiments and obtained similar results. The following are Riess' results.-(Pog. Ann. XLVII, 55.)

In the conducting circuit of the battery, consisting of 25 jars with 13 square foot coating each, a spiral of platinum wire was placed; 26 inches of this spiral were wound in 42 coils on a glass tube 3 inches long. The ends of the wire not wound up were, together, 34 inches long.

In each experiment a new non-magnetic English sewing needle, 13 9 lines long and 0.19 lines thick in the middle, was laid in the spiral. After the discharge stroke had passed through the spiral the needle was magnetized. To test the strength of the magnetism it was brought to a certain distance from a compass needle two inches long, (in what manner this was done cannot be easily understood from Riess' description,) and the deflection produced in the latter observed.

By increasing the charge of the battery, not only the strength but the polarity of the magnetism of the needle changed, as the following table shows:

Quantity, 5 10 15 20 25 27 29 30. 32 35 Deflection, 9° 14.5 15 10.3 6.5-2.5-7.5-8.5 2.3 11.5

It is seen that a stronger charge of the battery was not necessarily followed by a stronger magnetism; also, that the magnetism thus caused was not always such as might have been expected, according to Ampère's rule, (namely, that if we suppose the figure of a man to be introduced into the circuit, the positive current entering at the feet and passing out at the head, the figure, when it faces the needle, will have the north pole on its left hand,) for an abnormal magnetizing of the needle took place in all the deflections marked with the sign.

In this series the strength of the magnetism of the needle at first increased with the magnitude of the charge, then decreased until the direction of the magnetism was reversed, and it was only after still more powerful charges that the normal magnetism appeared again. These experiments are a proof that the direction of the discharge current cannot be deduced from the polarity of the needle.

With weaker charges the needle was normally magnetized; abnormal magnetism appeared with increased charges in fine needles only; coarse needles are always magnetized normally, although constantly increased charges produce in them an alternate decrease and increase of strength of the magnetism.

§ 58. Magnetizing by the secondary current.-This peculiarity in the magnetism of steel needles occurs in like manner in the secondary current. Magnetism produced by a secondary current will change in strength and direction:

1. By increasing the charge.

2. By increasing the surface of the battery, the charge remaining the same. The greater the surface, the stronger Riess found the magnetism of the needle; the same quantity of electricity being distributed

over a greater surface, it has a less density, and consequently a slower discharge, which is favorable to the production of magnetism.

3. The order of the periods of decrease and increase, as well as that of the reversal of the magnetism, will be changed by an alteration in the secondary circuit, such as introducing wires of constantly increasing length.

If the secondary circuit remains metallic as before, but interrupted at one place, so that the current has to pass with a spark, a very remarkable influence is observed on the magnetic effect; often the magnetism is in this way increased very greatly, sometimes it is weakened, and again it is changed in direction. The strongest magnetization by the secondary current, amounting nearly to saturation of the needle, has been obtained in this manner.

4. A continued change in the strength, as well as a change in the direction of the magnetism produced by the secondary spiral, takes place when, cæteris paribus, the length of the conducting circuit of the main spiral is continually increased.

The apparatus shown in figure 62 may be very conveniently used in these experiments. The lower spiral may be taken for the secondary circuit, and the magnetizing coil may be introduced between x and y by screwing its ends in the clamps.

§ 59. Production of heat by the secondary current.—It has already been mentioned that the secondary current produces thermal phenomena; Riess has also investigated thoroughly the laws of the development of heat by the lateral current.-(Pog. Ann., XLVII, 65.)

In the conducting circuit of the secondary spiral, a magnetic spiral · and an electrical air thermometer were introduced. The following table contains the thermal and magnetic effects which the secondary current produces when the surface and charge of the battery are: changed. S and q have the same signification as before.

In the last column the deflections of the compass needle produced by the magnetized needle are indicated as explained above. Where no deviation is indicated the magnetism was not perceptible.

As far as the last three observations, indicated by *, the observed temperatures harmonize very well with the formula

From all the observations (the tables given by Riess contain a few more) the mean result for a was 0.075; the temperatures computed with this co-efficient in the above formula accord perfectly well with the observed values. Hence the formula holds good for the temperatures produced by the secondary current.

*

In the observations indicated by the secondary circuit was interrupted, so that the current had to pass with a spark. This has a very important influence (above mentioned) upon the magnetization. It is shown here, while the heating power is scarcely affected-it being a little diminished.

When a German silver wire, 78 lines long and half a line thick, was inserted in the main circuit the heating was less; the co-efficient a, which was found above equal to 0.075, was now 0.028.

As may be readily conceived, the quantity of electricity in the secondary current is greater in proportion as the portion of the main spiral acting upon the lateral spiral is greater, other circumstances being equal. In order to determine the amount of increase of the secondary current thus produced, the secondary coil B B, (fig. 59,) closed by the platinum wire of the thermometer, was slipped over the straight prolongation of A A, and the temperature noted which was produced in the secondary wire by the discharge of q = 20 in 85. Then, in successive experiments, a different number of coils of the main spiral was brought under the secondary spiral, and the same quantity of electricity discharged in the same manner. These experiments gave the following results:

Length of straight No. of coils. Heating in the lateral

The numbers of the last column are the mean of two series of experiments, giving nearly the same results.

Since we know what elevation of temperature (1.85) is produced in the secondary wire by the action of a straight piece of the main wire 134 lines long, we can compute the heat produced by the action of a straight piece of the main wire 102, 634, &c., lines long, and thus