IV. Notice of a Chromatic Stereoscope. By Sir DAVID BREWSTER, K.H., F.R.S., V.P.R.S. Edin.* IN N the year 1848, I communicated to the British Association, at Swansea, a brief notice of the principle of this instrument +. If we look with both eyes through a lens, about 2 inches in diameter or upwards, at an object having colours of different refrangibilities, such as the coloured lines on a map, a red rose among green leaves, or any scarlet object upon a blue ground, or in general any two simple colours not of the same degree of refrangibility, the two colours will appear at different distances from the eye of the observer. In this experiment we are looking through the margin of two semilenses or virtual prisms, by which the more refrangible rays are more refracted than the less refrangible rays. The doublycoloured object is thus divided into two as it were, and the distance between the two blue portions is as much greater than the distance between the two red portions (red and blue being supposed to be the colours) as twice the deviation produced by the virtual prism, if we use a large lens or two semilenses, or by the real prisms, if we use prisms. The images of different colours being thus separated, the eyes unite them as in the stereoscope, and the red image takes its place nearer the observer than the blue one, in the very same manner as the two nearest portions of the dissimilar stereoscopic figures stand up in relief at a distance from their more remote portions. The reverse of this will take place if we use a concave lens, or if we turn the refracting angles of the two prisms inwards. Hence it follows, and experiment confirms the inference, that we give solidity and relief to plane figures by a suitable application of colour to parts that are placed at different distances from the eye. These effects are greatly increased by using lenses of highlydispersing flint glass, oil of cassia, and other fluids, and avoiding the use of compound colours in the objects placed in the stereo scope. * Read before the Royal Scottish Society of Arts, Dec. 10, 1849. + See Report of the British Association at Swansea, 1848, Trans. of Sect., P. 48. V. Account of Experiments with a powerful Electro-magnet. By J. P. JOULE, F.R.S. &c.* SOME years ago I announced that if a particle of wire conducting a voltaic current be made to act upon a very large surface of iron, the intensity of the induced magnetism will not be much diminished by an increase in the distance of that particle from the surface of the iron. Guided by this principle, I constructed a very powerful electro-magnet in 1843†, and soon after prepared the iron of the electro-magnet employed in the experiments related in the present paper. This was a plate of the best wrought iron, 1 inch thick, 22 inches long, 12 inches broad at the centre, but tapered thence to the breadth of 3 inches, as represented in the adjoining sketch (fig. 1). The plate was then bent into a semicircular shape, so as to bring its ends within 12 inches of one another. Previously to fitting up this bar as an electro-magnet, I made a few experiments with a view to test the principle above named more completely than I had hitherto done. A length of about eight yards of insulated copper wire, th of an inch in diameter, was divided into two exactly equal portions, one of which was wound four times round the broadest part of Figl the iron, and close to its surface; the other was also wound four times round the broadest part of the iron, but was kept at the distance of one inch from its surface by means of interposed pieces of wood. A constant current of electricity was alternately passed through the wires; and the deflections of a magnetic needle half an inch long, placed at the distance of two feet from the iron bar, were observed to be as follows: 6° 23' with the wire close to the surface of the iron. 6° 9' with the wire at the distance of one inch from the surface of the iron; showing only a trifling diminution of effect in consequence of the removal of the wire to the distance of one inch from the surface. Having been thus fortified in my previous conclusion as to the propriety of enveloping broad electro-magnets with a very large quantity of coils, even though the outer ones should be * Communicated by the Author. + Philosophical Magazine, S. 3, vol. xxiii. p. 268. removed to a considerable distance from the surface of the iron, I proceeded to fit up the large bar already described with a coil consisting of a bundle of copper wires 68 yards long, and weighing 100 lbs. The electro-magnet thus formed was placed in a wooden box, on the side of which two large brass clamps were screwed, the latter being soldered to the terminals of the coil. The accompanying sketch represents the apparatus in its comFig 2 pleted state; excepting, however, two brass straps, by means of which the coil is kept securely in its place, which are omitted for the sake of clearness. In experimenting with the electro-magnet, I employed a battery consisting of sixteen Daniell's cells, the copper of each exposing an active surface of nearly two square feet. They were arranged so that I could with facility use either one cell alone, four cells in a series of two, or sixteen in a series of four elements. The cells and the liquids in them being similar in every respect, it was evident that these arrangements must produce through the electro-magnetic coils currents represented by 1, 2 and 4. I therefore was enabled to dispense with the use of a galvanometer, which would have been acted upon by the powerful electro-magnet, even if it had been placed at the distance of many yards from it. Experiment I.-A magnetic needle, 1 inch long, was suspended at the distance of three feet from the electro-magnet measured on a line at right angles to that joining the poles. The northward tendency of the needle having been counteracted by means of a permanent magnet, I observed the following vibrations per minute resulting from the action of the electromagnet: The vibrations are evidently in the ratio of the square root of the quantity of current circulating around the electro-magnet, and consequently we may infer that the magnetism induced in the latter was simply in proportion to the current. Phil. Mag. S. 4. Vol. 3. No. 15. Jan. 1852. Experiment II.-Having provided a pair of tapered poles terminating in vertical edges, 1 inch long and th of an inch in breadth, I caused them to be slid on the poles of the electromagnet until within 1 inch from each other. A cylindrical bar of bismuth 13 inch long, of an inch in diameter, and weighing 174 grains, was suspended by a filament of silk from a proper support, so as to vibrate between the tapered poles. The average numbers of vibrations in each minute of time through the quadrant of a circle were then found to be The currents being as 1, 2 and 4, and the vibrations 41, 91 and 17, or nearly in the same ratio, it follows that the repulsive action of the magnetic poles was as the square of the current, and consequently that the diamagnetism of the bismuth is a quality not self-inherent, but induced by the magnetic action to which it is exposed. I am happy to have been thus enabled to confirm the important fact, discovered by M. Ed. Becquerel and Dr. Tyndall, by experiments made without any knowledge of the researches they were conducting almost simultaneously on the same subject. Experiment III.-The tapered poles remaining at 1 inch asunder, I suspended a piece of soft iron, 3 inches long, 1 inch deep, and th of an inch thick, at the distance of a quarter of an inch above the poles. Using one cell of the battery, this small piece of iron was attracted with a force of 63 oz.; but with 16 cells in a series of 4, with a force of no less than 71 oz. In this instance we notice a slight falling away from the theoretical attraction, owing no doubt to the gradual approach of the limit to magnetizability in the small bar of iron. Experiment IV.-The tapered poles having been removed, a flat bar of soft iron, 14 inches long, 3 inches in breadth, and 1 inch thick, was placed at various distances from the poles of the electro-magnet, and the attractions measured as follows: in. dist. in. dist. 1 in. dist. 2 in. dist. Here, again, we have evidences of an approach towards the limit *The electro-magnet with which the above experiments had been made was sent to the Exhibition of Industry in the middle of February. M. Becquerel's paper was published in the Annales de Chimie for May, Dr. Tyndall's in this Magazine for September, after having been previously communicated to the Ipswich Meeting of the British Association, of magnetizability, for the attractions with a current of 1 are only ten times, instead of sixteen times as great as those observed with a current of 1. The electro-magnet I described some years ago* consisted of a core of iron, half an inch thick, enveloped by a coil of wires weighing 60 lbs. With a battery of ten cells, similar to those employed in the present experiments, a bar of iron 3 inches broad and an inch thick, was attracted at the distance of of an inch with a force of 480 oz., at an inch with a force of 168 oz., and at 1 inch with a force of 77 oz. Both electro-magnets having been constructed on the same principle, their attractive powers ought to be proportional to the weight of coil and number of cells, and therefore to be represented by 60 × 10=600, and 100 × 16=1600. As this is tolerably well borne out by comparing the actual results of the above experiments, we may infer that little or no advantage was obtained by increasing the thickness of the core of iron from half an inch to one inch. Experiment V.-A flat bar of iron, 1 inch deep and th of an inch thick, being placed with its thin edge in contact with the poles of the electro-magnet, the following weights had to be applied in order to overcome the attraction in contact :— But when the bar of iron used in Experiment IV. was placed in contact with the poles, so as just to leave a quarter of an inch in breadth for the place of contact of the flat bar, the attraction of the latter with 16 cells was found to be only 82 lbs. Thus it would appear that 14 lbs. out of the 96 lbs. in the previous experiment were owing to the distant attraction of that part of the poles not in contact with the bar. We may therefore conclude, that while the attraction in contact, using one cell, was 64 lbs., minus say 1 lb. for distant attraction, that produced by a current four times as great was only increased to 82 lbs. And it must be remarked, that the greater part of this small increase was doubtless owing to the action of the broader part of the iron core which still remained unneutralized. It would therefore appear, that the greatest observed attraction in contact was, in this electromagnet, about 70 x 5=350 lbs. per square inch of the surface of each pole, or otherwise that the greatest magnetic attraction of one square inch of surface for another square inch was 175 lbs. Several years ago I gave 140 lbs. as the apparent limit of attraction in contact. The force of current employed in obtaining Philosophical Magazine, S. 3, vol. xxiii. p. 268. |