but it has been generally supposed that this involves instability in the ether. But Sir William, in the article alluded to, proves mathematically that this is not so if we regard the ether either as filling all space or as having fixed boundaries. Prof. J. Willard Gibbs, of Yale (" American Journal," February, 1889), admits that this theory is a formidable rival to the electric, but he still prefers the latter. Sir William also, despite his championship of the elastic theory, has publicly announced his conversion to the electric-brought about by the experiments of Hertz. Next in interest are the experiments of Hertz, William Hallwachs, E. Wiedemann, and others on the influence of light on the electric discharge. Under the influence of light rich in ultra-violet rays, the potential of a highly charged disk diminishes, and a current is even set up with other bodies in the vicinity. Borgeman showed the passage of a current directly between two flames. Hall wachs has shown experimentally that a negative charge diminishes more rapidly than a positive one, and A. Stoletow that a current is produced without a battery between gauze and a solid disk of more negative metal. Prof. Righi constructs photo-electric cells" of a disk and a net of different metals, placed opposite one another, and connected with an electrometer. When one is illuminated a deflection is obtained. J. Borgman interrupted the beam of light by a rotating perforated disk. No sound was heard in a telephone connected with the metallic conductor on which the intermittent beam fell, showing that the action of the light was not instantaneous, and therefore, as he claims, that the effect is secondary. The same experimenter (Paris Academy, April, 1889) has shown that the loss of negative electricity in the conductor increases with the time, and that it sometimes continues, still varying periodically, after the close of the illumination. 66 Light with Minimum Heat.-Prof. Brackett, of Princeton, in a lecture before the New York Electric Club, in January, 1889, discusses the possibility of producing luminous radiation unaccompanied by non-luminous rays. In the present method of lighting by incandescent solids, either suspended in flames or otherwise, the visible rays must always be attended by invisible heat rays, thus wasting for our purposes many times the available energy. The experiments of Hertz, already mentioned, lead us to hope that by attacking the problem from the electro-magnetic side the desired object may be obtained. The phenomena of phosphorescence and fluorescence show that light rays unaccompanied by dark ones are quite possible. If such radiation be obtained, Prof. Brackett thinks it possible to transmit it by wires from one place to another. Others think this impossible on account of the enormously increased resistance due to the rapid oscillations of the luminous electric wave. Spectroscopy. - Prof. Grünwald, of Prague, thinks he has established the principle that those wave lengths of light, in the spectrum of a substance A that belong to the element a are to those due to that element in the spectrum of a substance B as the atomic volume of a in A is to that of a in B (B being a compound of A with some other substance). This conclusion, which leads to important results-for example, the com 66 pound nature of hydrogen-is not accepted by physicists generally. C. Fievez, in a communication to the Belgian Royal Academy, offers a new interpretation of the spectral rays which regards them as due, in part, to interference. Hermann Ebert (Wiedemann's Annalen," 1889), claims that the width of spectral lines is much greater than can be explained by Doppler's principle applied to gaseous molecules, but Lord Rayleigh ("Philosophical Magazine," April, 1889) makes a fresh mathematical examination of the subject and concludes that this is not so, though there is much room for further discussion and experiment. Mr. E. F. J. Love has devised a new method of discriminating between real and accidental coincidences between lines of different spectra on the theory of probabilities. He has applied his test to some of the experiments of Prof. Grünwald, and finds that, so far as it goes, it sustains the latter's theory (see above). Prof. Samuel P. Langley, of Pittsburg, continues his researches on the energy of various parts of the spectrum. In a paper on "Energy and Vision" ("Philosophical Magazine," January, 1889), he says that the eye can perceive lights whose intensity varies in the ratio of one to one thousand million million. The time required to perceive faint light is one half second, but the time it takes the eye to recover its sensitiveness after exposure to a bright light is relatively long, being greatest for violet rays. The visual effect for the same amount of energy varies enormously with the wave length, being one hundred thousand times as great in the green as in the crimson. Prof. Langley's researches on the infra-red spectrum show that in this region the ratio between solar and lunar heat is completely changed, this ratio being five hundred thousand in the visible spectrum and five hundred in the infra-red. Wladimir Michelson, in a communication to the Russian Société Physico-Chemique, deduces from theoretical considerations the curves of energy of spectra that have been obtained experimentally by Prof. Langley. Among other laws, Mr. Michelson deduces the interesting one that the wave length that corresponds to the maximum energy is inversely proportional to the square root of the absolute temperature of the source. Edward Becquerel (Paris Academy, February, 1884,) has deduced the law that birefringent crystals have absorption spectra of different intensities in different directions, which in general coincide with those of the optic axes, but if two isomorphous substances having different optical properties are crystallized together, while the direction of the optic axes takes up a new position, the original direction of maximum absorption bands of each substance is retained, so that the absorption spectrum of each of the constituents can be observed. A powerful instrument for the analysis of crystalline substances is thus obtained. The selective absorption of metals for ultraviolet light has been observed at Harvard College by Prof. John Trowbridge and W. C. Sabine, who find that the color of the metal influences in no way its selective absorption for these rays. A new method in spectrum analysis has been devised by the same experimenters, who have used a steam jet in connection with the spark of a Leyden battery for the production of gaseous and metallic spectra. The jet impinges directly on the electrodes, and the resulting light resembles that of the electric arc. The steam is decomposed, giving the lines of hydrogen and oxygen, and those of the metallic spectra are much more distinct. Franz Stenger advances the theory that the absorption of light by substances depends primarily on the size of its physical molecules, and thinks that these are more complex in concentrated than in dilute solutions. Knut Ångström, of Stockholm (Wiedemann's "Annalen," March, 1889), in spectroscopic experiments on the quasi-absorption of light by small solid particles, finds that so long as the particles are of the same order of magnitude as the wave lengths of light, the change of transparency with increasing wave length is great, but if the particles are small in comparison with wave length, the medium has the properties of a homogeneous one that possesses real absorption. Prof. Henry A. Rowland, of Johns Hopkins University, has issued a new edition of his photographic map of the normal solar spectrum, of which the first appeared in 1886. These photographs were produced by concave gratings, 6 inches in diameter and of 214 foot radius, ruled with from 10,000 to 20,000 lines to the inch by a new dividing-engine, giving a definition far superior to any other that has yet been obtained. Prof. Rowland has devoted years to the study of dry plates, and has revised the list of standard wave lengths, carrying it into the ultra-violet. ence Wave Length of Light.-Louis Bell, of Johns Hopkins University ("American Journal of Scifor April and May, 1888), reports the conclusion of his experiments on this subject which were begun several years ago. The mean value of the wave length for the line D he finds to be 5,896-18 in air at 760 millimetres pressure and 20° C. temperature, or in vacuo 5,897.90, which he considers not likely to err by so much as one part in 200,000. C. C. Hutchins ("American Journal of Science," June, 1889) has determined the metallic spectra with greater precision than has been done before. Thalén's determinations (the best hitherto) are subject to errors of one in 3,000 or 4,000, while Rowland's recent determinations of solar lines are correct to one in 500,000. Mr. Hutchins used a 5x8 centimetre grating and an 8-inch spark, assisted by steam as in Trowbridge and Sabine's method (see above). His determinations make it probable that zinc exists in the sun, and almost certain that copper is there. Prof. Albert A. Michelson and Edward W. Morley ("American Journal of Science," September, 1889), discuss the establishment of a material standard whose length in light waves is known, and conclude that with a slight improvement in methods this can be done with an error of one part in a million, and perhaps within one in ten million. Dr. Hermann Ebert, of Erlangen, using eight different colored light sources, varying in intensity between 1 and 250, has established the constancy of the wave length to nearly a millionth, within these limits. Polarization.-Georges Meslin (Paris Academy Jan. 16, 1888), finds that when polarized light is transmitted through metallic films the rectilinear polarization becomes elliptical as in metallic reflection. Cornu (Paris Academy, May, 1889) has discovered by photographic registration of ultra-violet radiation that in that part of the spectrum the laws of elliptic polarization in vitreous and metallic surfaces approach each other. That is, if more and more refrangible rays be used, the coefficient of ellipticity in vitreous substances increases. J. L. Soret (Paris Academy, Nov. 26, 1889) finds that marine and lacustrine surfaces cause important perturbations in atmospheric polarization. Under certain conditions, he observed the phenomenon of two neutral points at the altitude of the sun, one on its right and one on its left. The polarization was in a vertical plane between these points and in the opposite direction beyond them. Rotatory Polarization.-In a discussion on this subject in the London Physical Society, May 25, 1889, Mr. A. W. Ward gave as his opinion that in magnetic rotation the periods of the light waves are altered, their velocity remaining the same. The effect in quartz he supposed due to the light itself, and not to the crystal, for liquids show the same. In opposition to this, Prof. S. P. Thompson mentioned the fact that fused quartz has not rotatory power. Prof. Thompson considers liquid rotation as due to some sort of "skew symmetry" of the molecules, the average effect of which is observed. Prof. Kundt (Berlin Physical Society, April 26, 1889), concludes, from experiments on glass made doubly refractive by pressure, that electro-magnetic rotation is common to all substances, but in doubly refractive crystals is a periodic function of the difference of phase of the two rays. A. W. Ward (London Royal Society, May 9, 1889), also arrives at the latter conclusion. Chauvin (Paris Academy, June, 1889), obtains similar results in the special case of Iceland spar, finding that in the direction of the axis there is simple rotation, and in other directions rotation accompanied by elliptical polarization. The rotation changes sense periodically and in certain directions disappears altogether, the elliptic polarization be ing alternately a maximum and zero in these directions. Lodge's Leyden jar experiments, described below, under "Electricity," show that the rotatory effect of an electric current is practically instantaneous, at least 70 of a second, whereas Villari, from experiments on a revolving glass drum, supposed it to take from to second. The results of Villari's experiments may have been due to strains in the revolving glass, as is concluded by A. W. Ward (London Royal Society, May 9, 1889) who repeated those experiments. Lodge attempts to explain magnetic rotation by hysteresis (q. v. under "Magnetism "), but A. Potier (Paris Academy, April, 1889) explains it by assuming that at each point of the medium ponderable matter tends to be carried along with a speed proportional to that of the ether in a light wave. Ponderable material molecules in the field become magnets, and being thus caused to oscillate induce electric force in the medium. Reflection.-Sir John Conroy finds that the amount of light reflected by polished glass varies with the method of polishing. After polishing, the surface of flint glass alters somewhat readily; that of crown glass slowly, the amount of reflected light decreasing and that of transmitted light increasing. A film of lower refract ive index appears to be formed. Dr. Rubens (Berlin Physical Society, March 8, 1889) has investigated the selective reflection of metals. He finds that the maximum reflexive power for silver is in the red, and for gold in the yellow. Copper reflects the blue and green rays less than gold, but increases rapidly toward the red and then more slowly into the infra-red spectrum. The reflexive power of iron and nickel rises rapidly at first from the blue toward the red, and then more slowly. In the infra-red it is not so high as copper or silver. The dispersions and indices of refraction deduced from these observations are similar to those of Kundt, now to be described. Refraction. In a communication to the Prussian Academy of Sciences, in February, 1888, Prof. Kundt describes his successful construction of minute prisms of seven metals, and his direct measurement of their refractive indices. The prisms were deposited electrolytically on platinized glass after several thousand trials. Kundt's results for red light were: 1.81 2.61 surfaces, by means of an ingenious color photometer of their invention (Royal Society of London, June, 1888). In a simpler apparatus (Physical Society, Nov. 24, 1888), two separate beams of light throw shadows of a rod, one on a white and the other on a colored patch. The amount of light falling on the latter is varied by a rotating disk with angular openings that can be altered while the disk is in motion. When the luminosity of the patches appears the same, a white patch is substituted for the colored one, and the comparison made again. The ratio of the size of the angular openings in the two cases gives the relative luminosity. Incandescence.-H. F. Weber's experiments cast doubt on the supposed fact that the dull red rays are the earliest to appear when a body is heated to incandescence. He found that carbon, platinum, gold, and iron give a "gray glow" at comparatively low temperatures. Interference.-Lord Rayleigh (" Philosophical Magazine," August, 1889) calls attention to the phenomena of achromatic interference bands, which, though known to Sir Isaac Newton and 217 treated of by Fox Talbot, have been lost sight of in recent times. The mathematical theory of bands shows that they will not be colored if the distance between the sources of light producing them vary with the wave length. This Lord Rayleigh brings about by shutting off all the spectra of a diffraction grating but one, which is reflected at a grazing incidence from a mirror whose plane passes through the central white spectrum of the grating. The distance of each ray of the admitted spectrum from this central image being proportional to the wave length, a succession of black bands is thus produced. In one case 1,200 lines to the inch were observed, a photographic image of which could be used as a diffraction grating. With a lens instead of a grating imperfectly achromatic bands were obtained. The velocity of light in these metals is inversely proportional to those numbers, and seems to be proportional to their conductivity for heat and electricity. Dr. John Kerr, of Glasgow, has continued his researches on the birefringent action of strained glass, finally specifying the wave surface in such glass, and establishing, among other conclusions, that the velocity of light along the line of strain, and at right angles to it, is diminished by compression and increased by ten sion. Photometry. The committee of the British Association on a standard of light, after testing six classes of standards for four years, recommended, in 1888, the pentane lamp (see also Definitions under "Electricity"). J. Joly has devised a photometer, which consists of two parallelopipeds of paraffin in contact, placed between the lamps to be measured so that the line joining the lights is perpendicular to the plane of junction. The instrument is moved until the line of division is no longer visible, when the relative intensity is calculated from the law of inverse squares. Drs. Lummer and Brodhun (Berlin Physical Society, Dec. 28, 1888) have invented a photometer consisting of two right-angled prisms with their hypotenuses in juxtaposition, a drop of Canada balsam being placed between. Then through the drop the light can pass in a straight line, while elsewhere total reflection cuts off all but side light. The lights to be compared are then placed, one behind and one at the side of the compound prism, and observation is made as with Bunsen's grease-spot photometer. As the drop of balsam loses its sharp edges, the inventors produce the same result by grinding to a slightly spherical form all of one surface but a spot in the middle. On pressing the surfaces together, light passes directly only through the central spot. The sensitiveness of this instrument is about 1 per cent. Capt. William de W. Abney and Gen. Festing, of the British army, have been able to measure the relative amounts of light reflected from colored and white or black Disintegration by Ultra-violet Rays.-Philip K. Lenard and Max Wolf (Wiedemann's " Annalen," 1889) find that the ultra-violet rays produce dust on the surface of negatively electrified metals and also on quartz and gypsum. The dust was detected by the condensation method of Aitken, described above. Radiant Energy of Flames.-Von Helmholtz (Berlin Physical Society, June 7, 1889) has measured the relation between the energy of flames and the amount of gas consumed. He finds that there is more luminous than non-luminous energy, and that the radiating power is not dependent on the temperature. His results are consistent with the hypothesis of Julius that the products of combustion are the only criteria of the amount of radiant energy of a flame. The author considers it more economical to use gas for driving a dynamo, which supplies an electric glow-lamp, than to burn the gas directly. Fluorescence.-B. Walter (Wiedemann's " Annalen," February, 1889) finds that the fluorescence of a quite concentrated solution of fluorescin is zero, or infinitely small. When this solution was diluted to, the fluorescence began to be measurable, and it then increased till the dilution reached, after which it remained constant for all degrees of dilution experimented with, the greatest being about 600000• The fluorescence, both of this substance and eosin, was found to increase with the temperature. Optical Teaching.-Prof. Silvanus P. Thompson disapproves of the ordinary division of optics into geometrical and physical, and thinks that the theory of lenses, etc., should be treated from the first by using the idea of waves. He has begun ("Philosophical Magazine," October, 1889) a series of papers in illustration of his method. Electricity. Its Nature.-Prof. Oliver J. Lodge, developing the views of Michael Faraday and James Clerk Maxwell, has presented what he considers the "Modern Views of Electricity" in a series of articles in "Nature," beginning in October, 1887, and since published in book form (London, 1889). According to his ideas, electricity is probably identical with the luminiferous ether. According to the views of his school, the phenomena of statical attraction or repulsion are due to a state of strain in the dielectric in the neighborhood of the so-called "charged bodies," and Prof. Lodge explains that the ether may be thus strained while retaining the properties of a fluid, because the strain consists only in the separation of its positive and negative components. The phenomena of conduction are similarly explained, the transfer of electric energy in the latter case taking place, according to his view, through the surrounding medium, and not through the conductor (see Route of Electric Force below). Magnetic phenomena are explained as the result of vortices or whirls in the ether, while waves in it are identical with waves of radiant heat and light; but these are not waves of mechanical distortion as in the ordinary theory, but disturbances such as take place in the neighborhood of a body that is rapidly charged and discharged (see above, under "Light"). Prof. Lindemann, however, inthe theory of physical phenomena that has been mentioned, suggests that the atoms of those bodies that we call electrified are merely vibrating in very much shorter periods than those that give rise to light waves. An indefinite number of such small waves would impinge at once on a molecule, and the author shows mathematically that the effect would be the same as if the body from which they proceed were "electrified." As the luminous and electrical vibrations differ in his view only in frequency, a body should be electrified by being made to approach a source of light by Doppler's principle. This may explain the electrification of the particles of a comet's tail. By the same principle, a particle that moves away from a source of electricity should appear luminous, which may throw light on the phenomena of Geissler tubes. Sources of Electricity. A thermo-magnetic generator was suggested in 1869, before the Royal Society of London, by Dr. Gore, who proposed to produce a changing magnetic field by heating and cooling the iron core of a coil. In August, 1887, Thomas A. Edison described to the American Association a "pyromagnetic generator," which was the first of such generators to be actually constructed. He used eight horse-shoe magnets in a circle, their poles pointing inward. Between the poles of each was a roll of thin laminated iron covered with asbestos and wrapped with wire. The apparatus is placed over a fur nace from whose heat it is partially protected by a half-disk of fire clay. This can be made to revolve, alternately heating and cooling each bundle of iron, with accompanying variations in the magnetic field, and the consequent induction of a current in the surrounding coils of wire. In January, 1888, M. Menges of the Hague exhibited several new forms of this kind of generator, the chief of which consisted of a Gramme ring, within which was a stationary electro-magnet. In the space between the latter and the ring is a zigzag ribbon of iron, which is heated at such points as to destroy the symmetry of the lines of force. In consequence the ring rotates, generating currents as in an ordinary dynamo. Ferdinand Braun, in the Berichte der Berliner Akademie (1888) describes experiments that show that when a nickel spiral is suddenly pulled out. an electric current is generated, and when it is compressed there is a current in the opposite direction. The effect ceases when the wire is annealed. In Wiedemann's "Annalen" (MaySeptember, 1889) Braun treats exhaustively of the currents thus produced, which he proposes to call deformation-ströme (deformation currents). The direction of the dilatation current (that produced when the spiral is pulled out) in a righthanded nickel spiral is opposed to the direction of drawing. Metals which act in this way Braun calls negative. If the metal be magnetized longitudinally, so that the drawn end is a south pole, the current is increased. The deformation current is comparable in amount with thermocurrents, and is slight in iron and steel. Braun explains it by supposing that deformation alters the magnetism of the metal, causing a "magnetic current" through it, and that this change of magnetization induces an electric current. Dr. Carl Langer and Ludwig Mond have devised what they call a "dry gas battery," consisting of a porous diaphragm of plaster of Paris soaked with dilute sulphuric acid, both sides covered with perforated platinum leaf, over which is a film of platinum black. The diaphragm is arranged to form chambers, through which hydrogen is passed on one side and air on the other. One element, with an effective surface of 120 square inches, has an electromotive force of one volt and a resistance of half an ohm. The electromotive force is decreased by transportation of the acid from one side of the diaphragm to the other, but this effect is prevented by interchanging the gases from time to time. Dr. Wolff (Berlin Physical Society, Feb. 22, 1889) applies the name of "oxygen elements" to galvanic cells of zinc, or zinc sulphate or chloride, with copper, silver, or iron, believing his experiments to show that the source of current energy in such cells is due to the combination of oxygen with the metal. M. Hein has investigated the value of magnesium as a positive element in batteries. The electromotive force is high, reaching 2:95 volt with a magnesium-carbon couple in bichromate solution, and 2.98 when the magnesium is in dilute sulphuric acid. The disadvantages are the cost and the high resistance of magnesium salts in solution. Change of Potential of a Voltaic Couple.Between May and August, 1888, Dr. George Gore described, in a series of papers to the Royal Society of London, his experiments on the mini mum amount of various soluble substances re- of gas that is condensed on it. C. II. Draper, in quired to alter the electromotive force of a couple. This he found varied with the chemical composition of the liquid, the kind of positive metal, to a less degree, with the kind of negative metal, the temperature, and the kind of galvanometer employed. Contact Electromotive Force.-C. V. Burton (London Physical Society, April 28, 1888) deduces theoretically the laws that for substances chemically inactive "the true contact E. M. F. is equal to their Peltier effect expressed in absolute measure," and for substances without Peltier effect the E. M. F. is equal to the energy of combination of one electro-chemical equivalent." M. Peltschikoff (Paris Academy, July 15, 1889) finds that the contact electromotive force of a crystal has different values, according as it is taken on the top, a face, or an angle; also, that if one of two bodies is not isotropic (for instance, if it has some sort of symmetry with respect to an axis), its electromotive force of contact with the other body has the same sort of symmetry. Prof. Herroun (London Physical Society, Jan. 26, 1889) concludes that the primary factor in the electromotive force of a cell is the relative heat of formation of the anhydrous salts of the two metals, but this may be, and usually is, supplemented by the energy due to their hydration or solution. His experiments have enabled him to correct the received values of the heat of formation of salts. Thermo-electricity.-Herbert Tomlinson has shown that when part of an iron circuit is twisted, and the junction of the twisted and untwisted parts is heated, a current passes from the strained to the unstrained metal, which suddenly increases in intensity when a red heat is reached. Similarly, though hot iron is always negative to cold iron, the difference of potential increases suddenly at a red heat. This confirms the conclusion that at a high temperature iron undergoes a sudden alteration in molecular structure (see also Recalescence, under " Heat "). Albert Campbell (Royal Society of Edinburgh, Jan. 16, 1888) shows that tin at its melting-point undergoes a change in its thermo-electric properties similar to that which takes place in iron at a high temperature while still solid. James Monckman (London Royal Society, May 31, 1889) finds that the thermo-electric properties of carbon are opposite below and above 250° C., as shown in the following table: his experiments on the polarization of platinum plates, concludes that the electromotive force of polarization depends on the current when this is below a certain value, and increases with the current more and more slowly, till finally an increased current has no further effect. Lightning and Lightning-Rods.--Observations on lightning, especially by photography, have been frequent of late. In lightning photographs, what appear to be images of dark flashes are often noticed on the plate. The cause remains in doubt. Prof. Stokes suggests that oxides of nitrogen in the track of a flash may cause the phenomenon by absorption of light. G. M. Whipple (London Physical Society, April 13, 1889) supposes it not to be a real phenomenon, but produced, in taking prints from the negatives, by successive reflection from the reduced silver and the glass. If this is so, a "dark flash' should always be parallel to a bright one. This is not always so, but this may be due to irregularity in the upper surface of the negative. Others think the effect due to some kind of reversion. For instance, A. W. Clayden (London Physical Society, June 22, 1889) supposes it due to exposure of the plate to diffused light from a cloud just after the flash. His experiments on the photography of an electric spark show that this is possible. Photography seems to indicate that several flashes often follow in the same path. This is also shown by the appearance of a mirror struck in Prague on June 9, 1889. The heated air in the track of one flash may serve as a conductor for the next. But G. M. Whipple supposes that the photographic effect is due to taking the pictures through glass windows, and he illustrates his point by so photographing a chalk line on a blackboard. Much evidence has also been collected on the subject of globular lightning, and it seems to be generally admitted that it exists, contrary to the opinion of early authorities, who thought it an optical illusion. Prof. Oliver J. Lodge, in a course of lectures before the Society of Arts, in 1888, disagreed with some received ideas about lightning. He asserted that a lightning discharge will often fail to follow the best conductor, and that there is a tendency to side-flash, the discharge often leaving the rod and following a very erratic path. According to his views, there are two principal causes of obstruction besides those that depend on the actual resistance-first, the direct effect of self-induction (called "impedance by Oliver Heaviside); and, second, the fact that a static discharge is rapidly alternating (see Leyden Jars). If the alternations are rapid enough, the current may be confined to the surface of the rod, producing a tendency to sideflash. For lightning-rods, he considers iron better than copper, its self-induction being less, although it is magnetic, perhaps because the current passes only on the outside, and hence magnetizes nothing. He thinks rods should be made of great capacity as well as high conductivity, and therefore recommends that the conductor be expanded over as much space as possible. This matter was afterward the subject of an interesting discussion in the British Association, where William H. Preece differed widely from Prof. Lodge's conclusions, saying that |