the wire equal to that of light; the other a greater velocity.

"Wireless Telegraphy." Marconi (“Nature," June 17) has experimented on a new plan of electric signaling through space, utilizing Hertzean waves of very high frequency, and hence employing conductors of very moderate length. He employs as a transmitter Prof. Righi's form of Hertz's radiator, and as a receiver a modification of Lodge's "coherer." With his apparatus signals have been sent nine miles. Weather has no influence and hills do not act as obstructions.

Cathode Rays.-J. J. Thomson (Cambridge [England] Philosophical Society, Feb. 8) has experimented on the electric charges carried by these rays and on their deflection in a uniform magnetic field. The result seems to show conclusively that there is a flow of negative electricity along the rays, but that there must be something else besides a stream of negatively electrified particles. He found that the magnetic deflection of the rays in various gases is the same, provided that the mean potential difference between the cathode and the anode is the same. Certain of the rays are not deflected by a magnet; these have little if any power of producing phosphorescence. The path of the rays for the first part of their course is approximately circular. Thompson (London Royal Society, June 17) finds that the size of the cathodic shadow depends on the electric state of the object that casts it, and that objects protected by a nonconducting layer of glass do not, at moderately low exhaustions, when made cathodic, repel or deflect cathode rays. He finds also that the rays can not be concentrated by reflection; that when they strike on an internal metal target, the latter emits other rays that resemble them in certain respects, but that can not generate Röntgen rays; and that metal screens sift out the rays, those most readily stopped being the ones most easily deflected by a magnet. Thomson and Skinner (Cambridge [England] Philosophical Society, Nov. 22) find that aluminum is rapidly evaporated from the cathode by an electric discharge in a highly exhausted vacuum tube in which mercury vapor replaces air. The metal is deposited on the glass in a bright layer which when dissolved off by acid leaves a gelatinous residue of silica. Other similar chemical effects are recorded by the same observers. Deslandres (Paris Academy of Sciences, June 8) finds that an electrified body interposed in the path of a cathode ray causes an enlargement of the shadow, the cathode bundle being divided up into several distinct and unequally deviated bundles. These, which he calls "simple cathode rays," correspond with simple electric oscillations. Bernstein (Wiedemann's "Annalen," November) asserts that the apparent repulsion between two parallel cathode beams arises from some action of the metallic cathode on the origin of the other beam and not to mutual action of projected particles. The paths remain straight when others proceed parallel from opposite ends of the tube. S. P. Thompson (British Association) distinguishes four kinds of cathode rays: (1) The ordinary kind; (2) those produced when the former fall on a surface and by producing X rays lose their power to excite X rays; (3) those that arise when ordinary cathode rays pass through a negatively charged metallic spiral or sieve (nondeflectable by a magnet); and (4) those that appear at the opening of a Holtz's funnel tube. These last produce no fluorescence and are magnetically nondeflectable. Tollenaar (Amsterdam Royal Academy of Sciences, Sept. 25) asserts that all the phenomena of two cathodes can be explained simply by supposing the rays to consist of negatively charged particles emitted with great velocity. Previously (ibid., Jan. 2) he had

Mc

found that by using as cathodes two square aluminum plates whose distance can be varied, the phosphorescence figures on the wall of the globular screen consisted of a zone of very intense green, bordered on either side by two rings. By changing the intensity of the current toward one cathode one ring obeyed the rules of deflection figures given by Goldstein, but the other behaved quite differently. With a triangular or square plate and a metal globe remarkable shadows were obtained. Clelland (London Royal Society, April 8) concludes that Lenard rays are simply a secondary propagation of cathode rays, produced by the rapid pulsations of negatively charged particles up to the aluminum screen, and that, like the cathode rays, they are streams of charged particles. Ròiti (“Atti dei Lincei," VI, page 5) has conducted experiments to ascertain whether Röntgen rays exist in the cathodic pencil that produces them, and concludes that in deflectable cathode rays either are nonexistent or not transformable into Röntgen rays. His experiments also indicate that metals of greatest atomic weight emit the most intense rays. Röntgen Rays. Their Nature. (See article on RÖNTGEN RAYS in "Annual Cyclopædia," 1896, and Cathode Rays, just above.) Thompson, in his inaugural presidential address before the newly formed Röntgen Society (Nov. 5), gave a review of recent progress with respect to Röntgen radiation ("Nature," Nov. 11). He said that the nonhomogeneousness of the rays was now generally admitted. Precht, of Heidelberg, in a recent thesis, reported in "Nature,” March 4, confirms the view that cathode rays and Röntgen rays are highly complex. He considers that part of such radiation is not a wave motion, but may be electric in nature. From interference phenomena the wave lengths of Röntgen rays were found to range from 370 × 10-6 to 830 × 10-6 millimetres. Some waves were four times as great as those obtained by Violle by diffraction through a slit. Vosmaer and Ortt, of the Hague ("Nature," Aug. 5), regard Röntgen rays as discharged cathode rays. The strongest proof for this theory, they say, is Lafay's experiment in which Röntgen rays that have passed through a negatively charged leaf of silver can again be deflected by a magnet in the same direction as the cathode rays in the tube, and, when the leaf is positively charged, in the opposite direction. Imbert and Bertin-Sans find that after prolonged use a Crookes tube emits rays that appear to differ from ordinary X rays in being able to traverse, without appreciable absorption, bodies relatively opaque to the latter.

Their Production.-Trowbridge, in a paper on "The Energy Conditions necessary to produce the Röntgen Rays" ("Proceedings of the American Academy," April), says that the discharge in a Crookes tube when on the point of emitting the rays most intensely is oscillatory, and that each discharge encounters a resistance of less than five ohms. Swinton ("Nature," Jan. 7) notes that Röntgen-ray tubes should be excited during exhaustion with a coil of the same dimensions as the one that they are ultimately destined to be worked with. Voller and Walter (Wiedemann's "Annalen," May) find that as exhaustion is increased less heat is developed in a discharge tube, the production of heat giving way before that of Röntgen rays.

Properties.-Battelli ("Nuovo Cimento," Vol. IV) and Villard ("Bulletin de la Société Française de Physique," 1895) have investigated the reflection of cathode and Röntgen rays. Both find that cathode rays are not regularly reflected, but that when they impinge on a thin metallic lamina rays having all their properties emerge from the opposite face, generally normally to it. Wind (ibid.) concludes that

vin (ibid., June 7) also finds that the efConductivity induced in air by uranium test in its immediate neighborhood, elecThe same kind as that of the uranium will ted in the air near it, and the opposite the inclosing metal surface. Hence, if s and shapes are such that for small e electricity near the uranium preponderYress must come to a maximum and then with greater and greater potential differwith still higher potential difference arge may exceed, and so the electricity may be of opposite sign to that found at results correspond with those found 1y by Dr. J. C. Beattie (ibid.). Becqueademy of Sciences, April 12) finds of electricity due to uranium appears solely by the gas in contact with the n ibid., April 26) claims to have es*he action of uranium is only a para general law that all substances n of light emit rays that discharge

in Gases.-Villari (Paris AcadeMarch 15) finds that gases that property of discharging electriby having been sparked or travrays, lose this power when subdischarge of an ozone apparatus. #ralization persists in the ozor it has been disconnected from accumulated charges on the experimenter (Naples Royal ed the property of cooling ed by gases after being travas, by observing the action of red-hot platinum spiral. In nt cooling produced a fall of ent. in the wire. The effect gy of the sparks, and at first of the spiral, but after this the refrigerating power deambridge [England] Philofinds that gases given off I reactions have sometimes ge, which they retain in a when bubbling through a in sulphuric acid. These Then they get into a moist the cloud being proporge, but heavier when the .nie (London Physical t when a V-tube containch, and the same metal, ected to a galvanometer of the curves that co-orectro-motive force deor heating of the metth. The effect is atStalline structure, and the melting point. t (Wiedemann's "Anat fluorspar, though it Some parts under the tive charge in other charges under the n shows no signs of Elster and Geitel elation between the cathodic absorption dence and direction ght are varied, find pends on these facit of light absorbed onfirm the accepted

Dielectrics.-Abegg (ibid., January) finds that as the specific inductive capacities of all substances increase with fall of temperature, the high dielectric constant of water may be approached by cooling other dielectrics to low temperatures. Hopkinson and Wilson (London Royal Society, Jan. 21) find that for long periods residual charge diminishes with rise of temperature in the case of glass, but for short times it increases, both for glass and for ice. The capacity of glass for ordinary durations of time (18 to 1% second) increases much with rise of temperature, but for very short times

1

second

it does not increase sensibly. This is due to residual charge. Conductivity has been observed in glass at fairly high temperatures and after brief electrification; after 0000-second electrification it is much greater than after electrification for 1000 second, but for longer times it is sensibly constant. Thus continuity has been established between conduction in dielectrics that exhibit residual charge and that in ordinary electrolytes.

Condenser.-Pollak (Paris Academy of Sciences, June) constructs a condenser of very high capacity by passing first an alternating current and then a direct current between aluminum poles in an alkaline solution. The plates become coated with a thin deposit of nonconducting oxide, which acts as the dielectric.

The Arc.-Wilson and Fitzgerald (“" Astrophysical Journal," February), in an investigation of the change of temperature of the arc as dependent on pressure, conclude, althougn some observations showed that the temperature was lowered with pressure, and although none showed an appreciable increase, that they can not affirm, on present evidence, that pressure either raises or lowers are temperature. Blondel (Paris Academy of Sciences, July 19), by experiments in which the passage of a current between carbon poles was broken for only ou second, during which the back electro-motive force of the poles was opposed to a single cell, eliminated the effect of the cooling of the carbons, and concludes that the arc behaves like an ordinary resistance, presenting no counter electro-motive force comparable to the potential difference of the carbons. Foley (American Association) finds that the electric arc has concentric layers of different color, the outer one being strongest in yellow. The upper carbon, whether positive or negative, is covered deeply with the yellow flame. Inside is a blue area, stronger at the positive carbon, and here, within the blue, a violet region is found.

Coherence.-Appleyard (London Physical Society, March 26) has devised some new experiments on coherence in liquid conductors. When mercury and paraffin oil are shaken together until the mercury divides into small spheroids, coherence of the latter can be visibly effected by a current, a spark, or a Hertz oscillator, the resistance falling from several megohms to a fraction of an ohm. When an emulsion of water and paraffin oil is formed the suspended globules of water are similarly precipitated by electrification, producing an artificial thundershower.

Nature of a Current.-Lord Armstrong, in a work entitled "Electric Movement in Air and Water" (London, 1897), concludes that an electric current consists of two streams, a negative one in the form of a core flowing in one direction, surmounted by positive electricity flowing in the opposite direction, and he suggests instead of “negative" and positive" the terms "inward” and "outward as more appropriate.

Hall Effect.-Van Everdingen (Amsterdam Royal Academy of Sciences, April 21) finds that the ratio of the resistances of bismuth in different directions

X-shadow figures obtained by Fomm and others are not ordinary diffraction images, but secondary ones. Analogous phenomena can be shown with ordinary light, and this fact, the author believes, goes far toward proving the X rays to be undulaatory. Haga (Amsterdam Royal Academy of Seiences, April 21) describes a new method of determining the wave length of Röntgen rays, the result of experiments by Tiddens, of Gröningen, founded on the fact that a perfectly identical deflection image is produced by rays of very different wave lengths, by varying the distance between source, diffracting slit, and screen in a definite manner. Benoist (“Bulletin de la Société Française de Physique," 1891) finds that for ordinary rays the specific absorbing power for the rays (defined as the absorption of a layer of unit surface density) increases with the density, but that results vary with the quality of the rays. He assumes that at the limit, as rays of higher and higher frequency are produced, the specific absorbing powers of all substances are equal. Swinton (London Royal Society) has investigated some properties of Röntgen rays by means of the luminescence they produce on a carbon surface. He has thus demonstrated the hollowness of the convergent and divergent cones of rays in a focus tube, has shown that the rays cross at the focus with no rotation, and has proved that the cathode rays, to produce Röntgen rays, must fall on solid matter. Hemptinne (Paris Academy of Sciences, Sept. 6) finds that while electric vibrations cause gases to become luminous at a low pressure, the Röntgen rays, used at the same time, raise the pressure at which luminosity occurs.

Effects.-Perrin (Paris Academy of Sciences, March 1) finds that the effect produced on a charged conductor by the Röntgen rays consists of two parts, one depending on the nature of the gas and the other on the metal. Villari (“Atti dei Lincei," VI, page 1) has investigated the relation between the dissociation produced in gases by Röntgen rays, in virtue of which such gases discharge electrified bodies, and the molecular association in which oxygen is transformed into ozone by the electric spark. He finds that an ozonator through which passes a current of air first traversed by Röntgen rays deprives that air of its power of discharging an electroscope. Hemptinne ("Zeitschrift für physikalische Chemie," December, 1896) has attempted to detect some action of Röntgen rays on chemical processes. He could detect no effect at all on the conductivity of aqueous solutions of electrolytes, the hydrolysis of ethereal salts by acids, and the combination of chlorine with hydrogen and carbon monoxide, and only minute traces of change on solutions, such as that of silver nitrate in alcohol, that are decomposed by light. Perrin (Société Française de Physique. Dec. 4, 1896), in investigations on the dissociation of neutral electricity in gases by Röntgen rays, finds that at constant pressure the quantity of dissociated electricity per unit volume is proportional to the pressure, and hence to the density. At constant pressure it is independent of the temperature. The quantity dissociated per unit mass varies as the absolute temperature. Child (American Association) finds that electricity can be continuously discharged by X rays even when the charged body is covered with a solid insulator, as paraffin. Villari (Nature," May 27) finds that the discharge of a conductor in air, provoked by Röntgen rays, seems to take place by convection, as if by an electric dance of the particles in air, roused by radiation. This discharge lessens when the exposed surface is diminished, as by covering it with paraflin. The paraffin, under the action of the rays, does not gain in conductivity. India rubber and insulating liquids behave almost

as paraffin. The conductive power of gases under the influence of the rays increases with their densitv. Guggenheimer (Paris Academy of Sciences, Feb. 15) shows that the Röntgen rays influence the explosive distance of the electric spark, the increase of this distance depending on the intensity of the rays that fall on the spark gap when the sparking distance and potential difference are constant. If the potential and the intensity of the rays are kept constant, the increase of explosive distance depends on the distance of the spark gap from the emissive wall of the tube.

Radiography by Impact.-De Heen ("Bulletin of the Belgian Academy") has obtained a photographic impression of lycopodium powder that is allowed to fall on a sensitive plate, by means of the electricity developed by the impact of the powder with the plate. With an uncovered plate a feeble but distinct impression was obtained after an hour and a half, but with a covered plate the impression was more powerful. Where the covering was broken, dark ramifications extend into the covered portion, apparently following the directions in which electricity was propagated along the surface of the plate. Anson ("Nature," Jan. 21) has secured very perfect images of the invisible electrie discharge. He says: This discharge, or possibly, more strictly speaking, the electrified streams of air driven off by it, appears to act upon the plate exactly as light does. It is thus possible to secure impressions of such discharges by simple electrification and subsequent development."

66

Discharge Rays.-Hoffmann (Wiedemann's "Annalen," February) describes a kind of rays called by him "discharge rays," which are contained in the spark discharge in air, hydrogen, and nitrogen, at ordinary or low pressures. They have no photographic action, but impart luminescence to solid solutions of magnesium sulphate in gypsum when heated to a temperature below incandescence. They are intercepted by mica, quartz, fluorspar, and other solids (unless produced at low pressures), proceed in straight lines, are not deflected in a magnetic field, and are not reflected by solids. They differ from ultra-violet light in their power of penetrating air and not fluorspar.

Radiant Properties of Uranium — Becquerel (Paris Academy of Sciences, March 1), in researches on his "uranic rays" (" Annual Cyclopædia,” 1896, page 641), finds that uranium has the property of discharging electrified bodies at a distance, the time of discharge being the same for positive and negative charges. For potentials below 15 volts the velocity of discharge is proportional to the potential. For potentials between 1,500 and 2,000 this velocity is constant. Lord Kelvin and others (Edinburgh Royal Society, March 1 and 9) obtain similar results. They find that two polished metallic surfaces connected to the sheath and the insulated electrode of an electrometer give a deflection from the metallic zero of about the same amount when the air between them is influenced by the uranium rays, as when the metals are connected by a drop of water. Kelvin, Beattie, and De Smolau (ibid., April 4) find that the electric leakage induced by metallic uranium is not proportional to the electromotive force, and is not perceptibly increased when the uranium is heated or when sunlight falls on it. The rate of leakage is greater in oxygen than in air, and less in hydrogen. In carbonic acid it is less for low and greater for higher pressures. In air at pressures from 760 to 23 millimetres the leakage is very nearly proportional to the atmospheric pressure. Lead 2 millimetres thick is not transparent to the influence of the uranium, but glass 3 millimetres thick does not stop it entirely. Mica is quite transparent, and copper and zinc partially

so. Kelvin (ibid., June 7) also finds that the effective conductivity induced in air by uranium being greatest in its immediate neighborhood, electricity of the same kind as that of the uranium will be deposited in the air near it, and the opposite kind hear the inclosing metal surface. Hence, if dimensions and shapes are such that for small voltage the electricity near the uranium preponderates, the excess must come to a maximum and then diminish with greater and greater potential difference, while with still higher potential difference the outer charge may exceed, and so the electricity drawn off may be of opposite sign to that found at first. These results correspond with those found experimentally by Dr. J. C. Beattie (ibid.). Becquerel (Paris Academy of Sciences, April 12) finds that the loss of electricity due to uranium appears to be effected solely by the gas in contact with the metal. Le Bon (ibid., April 26) claims to have established that the action of uranium is only a particular case of a general law that all substances under the action of light emit rays that discharge electrified bodies.

Other Effects in Gases.-Villari (Paris Academy of Sciences, March 15) finds that gases that have acquired the property of discharging electrified bodies, either by having been sparked or traversed by Röntgen rays, lose this power when subjected to the silent discharge of an ozone apparatus. This power of neutralization persists in the ozonizer for a time after it has been disconnected from the coil, owing to accumulated charges on the glass. The same experimenter (Naples Royal Academy) has studied the property of cooling heated bodies, acquired by gases after being traversed by electric sparks, by observing the action of different gases on a red-hot platinum spiral. In some cases the apparent cooling produced a fall of resistance of 10 per cent. in the wire. The effect increases with the energy of the sparks, and at first with the temperature of the spiral, but after this exceeds a certain limit the refrigerating power decreases. Townsend (Cambridge [England] Philosophical Society, Feb. 8) finds that gases given off during certain chemical reactions have sometimes a large electrostatic charge, which they retain in a striking manner, even when bubbling through a liquid or shaken up with sulphuric acid. These gases condense a cloud when they get into a moist atmosphere, the weight of the cloud being proportional to the electric charge, but heavier when the charge is negative.

Thermoelectricity.-Burnie (London Physical Society, Feb. 26) finds that when a V-tube containing solid metal in one branch, and the same metal, molten, in the other, is connected to a galvanometer by copper wires, the slope of the curves that co-ordinate temperature and electro-motive force depends on the rate of cooling or heating of the metal, particularly with bismuth. The effect is attributed to variation of crystalline structure, and is particularly marked just at the melting point. Photo-electricity.-Schmidt (Wiedemann's "Annalen," November) finds that fluorspar, though it acquires a positive charge in some parts under the influence of light and a negative charge in other parts, dissipates only negative charges under the action of light. Even selenium shows no signs of dissipating a positive charge. Elster and Geitel (ibid.), in experiments on the relation between the photo-voltaic current and the cathodic absorption of light when the angle of incidence and direction of polarization of the incident light are varied, find that the current, so far as it depends on these factors, is determined by the amount of light absorbed at the cathode. Their results confirm the accepted theory of metallic reflection.

Dielectrics.-Abegg (ibid., January) finds that as the specific inductive capacities of all substances increase with fall of temperature, the high dielectric constant of water may be approached by cooling other dielectrics to low temperatures. Hopkinson and Wilson (London Royal Society, Jan. 21) find that for long periods residual charge diminishes with rise of temperature in the case of glass, but for short times it increases, both for glass and for ice. The capacity of glass for ordinary durations of time (10 to 1 second) increases much with rise of temperature, but for very short times

1 106

second

it does not increase sensibly. This is due to residual charge. Conductivity has been observed in glass at fairly high temperatures and after brief electrification; after 0800-second electrification it is much greater than after electrification for 10000 second, but for longer times it is sensibly constant. Thus continuity has been established between conduction in dielectrics that exhibit residual charge and that in ordinary electrolytes.

Condenser.-Pollak (Paris Academy of Sciences, June) constructs a condenser of very high capacity by passing first an alternating current and then a direct current between aluminum poles in an alkaline solution. The plates become coated with a thin deposit of nonconducting oxide, which acts as the dielectric.

The Arc.-Wilson and Fitzgerald ("Astrophysical Journal," February), in an investigation of the change of temperature of the arc as dependent on pressure, conclude, althougn some observations showed that the temperature was lowered with pressure, and although none showed an appreciable increase, that they can not affirm, on present evidence, that pressure either raises or lowers are temperature. Blondel (Paris Academy of Sciences, July 19), by experiments in which the passage of a current between carbon poles was broken for only

second, during which the back electro-motive force of the poles was opposed to a single cell, eliminated the effect of the cooling of the carbons, and concludes that the arc behaves like an ordinary resistance, presenting no counter electro-motive force comparable to the potential difference of the carbons. Foley (American Association) finds that the electric arc has concentric layers of different color, the outer one being strongest in yellow. The upper carbon, whether positive or negative, is covered deeply with the yellow flame. Inside is a blue area, stronger at the positive carbon, and here, within the blue, a violet region is found.

Coherence.-Appleyard (London Physical Society, March 26) has devised some new experiments on coherence in liquid conductors. When mercury and paraffin oil are shaken together until the mercury divides into small spheroids, coherence of the latter can be visibly effected by a current, a spark, or a Hertz oscillator, the resistance falling from several megohms to a fraction of an ohm. When an emulsion of water and paraffin oil is formed the suspended globules of water are similarly precipitated by electrification, producing an artificial thundershower.

Nature of a Current.-Lord Armstrong, in a work entitled "Electric Movement in Air and Water" (London, 1897), concludes that an electric current consists of two streams, a negative one in the form of a core flowing in one direction, surmounted by positive electricity flowing in the opposite direction, and he suggests instead of “negative" and "positive" the terms “inward" and "outward as more appropriate.

Hall Effect.-Van Everdingen (Amsterdam Royal Academy of Sciences, April 21) finds that the ratio of the resistances of bismuth in different directions