Hydrodynamics. - Prof. Osborne Reynolds (British Association, 1894) has studied the successive stages in the motion of water passing under gradually increasing pressure through a vertical tube constricted in the middle. The water first leaves the constriction as a narrow steady jet, then it fills the lower part while eddies appear below the constriction, then the motion becomes turbulent, and finally there is an appearance as of air bubbles at the constriction with a singing or hissing sound. Science, Oct. 1) shows that a crystallizing body loses heat in essentially different ways, according as it is adiathermanous or diathermanous. All substances become diathermanous below-70° C., hence the true temperature of crystallization is obtained only by keeping the surrounding medium slightly below the solidifying point. This explains various anomalies, as in previous determinations of the crystallization points of chloroform, etc. Moore ("Zeitschrift für Physikalische Chemie, December, 1893) has measured the velocity at which crystallization proceeds in Decomposition of Liquids by Powders.-Dr. a supercooled substance, by following with the G. Gore (Birmingham Philosophical Society) eye the moving line of demarcation between finds that when finely powdered substances, essolid and liquid in a U-tube. With acetic acid pecially silica, are placed in a solution the adherthe velocity was found to be uniform at any tem- ent film of liquid contains more than the normal perature and independent of the diameter of the percentage of the dissolved substance. The tube, and with this substance, phenol, and mix- amount of the substance thus abstracted detures of phenol with water and phenol with cre- pends on the kind of powder, its fineness, its sol, the velocity increases with the amount of quantity in proportion to the dissolved subsupercooling, but at a diminishing rate. For stance, the absolute amount of the latter, and in phenol the velocity is 0·6 cm. a second with 4·4° some degree on the temperature. The results supercooling, and 2.9 cm. with 15.8°. The addi- seem to throw light on the purification of water tion of water or cresol reduces both the velocity by filtration. and the rate of its increase. Cohesion. Kasterine (“Journal of the Russian Physico-Chemical Society," xxv, p. 51), from experiments on the variation of cohesion in liquids, and on the assumption that the molecular forces conform to Newton's law of the square of the distance, derives the following laws: 1. The product of the intensity of molecular action by the molecular weight is a constant. 2. For different liquids at corresponding temperatures the radii of the spheres of molecular activity are approximately proportional to the square roots of the molecular weights. 3. At corresponding temperatures the physical molecule in different liquids contains the same number of chemical molecules. Density.-Kohlrausch and Hallwachs (Wiedemann's Annalen," October) have measured the density of very dilute aqueous solutions to within one millionth by weighing a glass globe suspended in the solution by a fine fiber. Interesting details regarding the molecular volumes of the dissolved substances have thus been brought out-for instance, phosphoric and sulphuric acids show a diminution of this volume at extreme dilutions. Capillarity. Quincke (British Association, 1894), as a result of researches occupying forty years, finds that drops of oil floating on slightly alkaline water are attracted toward the walls of the vessel and then repelled, the spreading of the soap film that results from the action of the alkali on the oil giving rise to periodic vortex motion. Viewed with the microscope the film shows the same minute strings of pearly beadlike bubbles that are characteristic of protoplasm. Viscosity-Owen Glynn Jones (London Physical Society, Feb. 9) has measured the viscosity of liquids by observing the velocity of small drops of heavier liquids falling through them. A water drop of 1 mm. radius was found to fall 1 inch an hour in castor oil at 8° C. This method has been criticised, the formulæ used being calculated originally for solid spheres, and the critics holding that internal motion in the drops as well as mutual contamination of the two liquids would complicate matters. Gases. Condensation.-In a lecture at the Royal Institution, published in “Nature,” Dec. 28, Shelford Bidwell states his conclusion from his own experiments, that the dense condensation of steam produced under certain circumstances is due neither to electrical action nor to dust nuclei, but probably to dissociated atoms of atmos pheric gases acting in some unexplained manCarl Barus (Nature," Feb. 15) objects to Mr. Bidwell's conclusions, and is of the opinion that condensation upon minute particles in the air will account for all the observed phenomena. ner. Critical Point.-Kuenen (Amsterdam Academy, May 26) has carefully investigated abnormal phenomena near the critical point, and concludes that they are due to impurities. From a repetition, for instance, of Galitzine's experiments, which have been supposed to show that ether above the critical point has different densities, according to its having been entirely fluid or partly vapor, Kuenen concludes that Galitzine's results were due to a trace of some noncoercible gas, perhaps air. He now Vortex Motions in Air.-Quincke announced in 1890 that two spheroids of a mixture of oil and chloroform, falling simultaneously in water, approached and receded alternately, owing to vortex motion in the water. finds (Wiedemann's "Annalen," July) that the same phenomenon occurs when two soap bubbles filled with coal gas ascend together. Similar phenomena occur when small dust particles fall in air, or liquid, or when a current strikes particles at rest. Barometry.-Bartrum has invented what he calls an * open-scale barometer," which is described in Nature," March 22, p. 488. The lower part resembles that of an ordinary mercury barometer, but near the upper surface of the mercury the tube is enlarged into a bulb, above which it is again contracted. The bulb is filled with a red fluid, the upper surface of which gives the barometer reading, a small change of level of the mercury causing a large one in the fluid. An inch of mercury is represented on the scale by 9 inches, and it is claimed that the atmospheric pressure may thus be ob tained to the thousandth of an inch of mercury without a vernier. Heat. Thermometry.-Barillé (Paris Acad-. emy of Science, Jan. 29) has devised an electricalarm thermometer for laboratory ovens. When the mercury reaches the desired point a circuit is closed by a platinum wire attached to a small iron tube that slides along a fixed wire and that is regulated in position by a magnet attached to the supporting frame. Baly and Chorley ("Nature," April 5) have devised a high-temperature thermometer in which the expanding substance is a liquid alloy of potassium and sodium. This alloy boils at about 700 C. and solidifies at -8° C., and the thermometer should be particularly useful for determining high boiling points. Specific Heat.-Silvio Lussana ("Nuevo Cimento," abstracted in "Nature," Sept. 20) has obtained the specific heats of gas by an ingenious apparatus that uses the same gas over and over again, and thus obviates the necessity of a large amount of gas. Expansion.-Max Toepler (Wiedemann's "Annalen," October) finds that the coefficient of expansion of elements in the solid state and their change of volume during melting show a definite relation. Conductivity.-I. T. Osmond ("Physical Review," November-December) finds that the thermal conductivity of cast iron is about 15 per cent. greater from a little below 100° C. to 200° than it is between 60 and 90°. Saret ("Bibliothèque Universelle," No. 4, 1893), from examining the shape of the isotherms in certain sections of crystals of gypsum, finds no evidence of the existence of rotational coefficients of conductivity. High Temperature.-C. Barus ("American Journal of Science," October) points out various anomalies in the accepted results of researches on high temperatures, and states that to clear these away either the boiling point of zine must come down from 930° to 905° C., or else the melting points of gold, silver, and copper must move up 30° or 40°, or both must move toward each other by corresponding amounts. Sound. Velocity.-J. W. Lowe (Wiedemann's "Annalen," August) finds, from experiments with a Quincke interference tube, that contrary to the results of Kundt, Regnault, König, and others-for closed tubes the velocity of sound in air and in carbolic acid is the same for notes of different pitch and intensity when they are propagated in open space. Pitch.-F. Melde (Wiedemann's "Annalen," June) has devised a new method of determining the pitches of high tuning forks independently of the ear. The fork is vibrated with one prong touching a metallic rod clamped at one end, which is thus caused to vibrate transversely, and the pitch is determined from the nodes of this rod, observed by means of fine sand that has been dusted on it. Beats. It is well known that two tuning forks produce beats even when one is held to each ear in such manner that a sound wave can not pass from one to the other through the air. Wundt thought that in this case the beats had their origin in the brain. Schaefer, however ("Zeitschrift für Psychologie und Physiologie der Sinnesorgane "), regards the effect as due to conduction by the bones of the skull. Residual Sensation.-Alfred M. Mayer ("American Journal of Science," January) has investigated the connection of the pitch of sounds with the length of their residual sensations—that is, with the sound perceived by the ear after actual vibration has ceased. The duration of the residual sound was measured by rotating perforated disks opposite the nipples of resonators that were sounded by tuning forks and noticing at what speed of rotation the interrupted sound seemed to become continuous. The duration was less the higher the pitch, varying from 00231 second for a frequency of 128 to 0.0049 second for a frequency of 1,024. Minimum of Audibility.-Lord Rayleigh ("Philosophical Magazine," September), in experiments to determine the minimum current audible in the telephone, finds that the maximum sensitiveness to currents occurs in the region of frequency 640, where a current of 44 X 10-8 ampères produced an audible sound. Telephones varied greatly in sensitiveness. The same writer (ibid., October), in experiments on the amplitude of just audible aërial waves, finds that for a frequency of 256 this amplitude is about 1.27 X 10-7 centimetres. Light. Luminosity.-P. Glan (Wiedemann's "Annalen," March) finds that the ratio of the volume of a candle flame to its illuminating power is very nearly constant, the difference between the actual luminosity and that calculated from this ratio being never greater than 3 per cent. Equal volumes of the bright flame of any two candles thus give the same amount of light. Reflection.-W. Wernicke (Wiedemann's" Annalen," March) shows that when light is reflected from a silver film between two transparent media, the anterior of which has the higher refractive index, there is not only a normal acceleration of phase increasing continuously from zero toor of a wave length as the thickness of the silver increases from zero to opacity, but also an anomalous retardation, which shows itself when there are traces of another substance between the silver and the front medium, and which may amount to between and of a wave length. Absorption.-G. B. Rizzo (Turin Academy) finds that Kirchhoff's law connecting the absorptive and emissive powers of substances does not hold good for cobalt glass. While its emissive power decreases nearly uniformly between wave lengths 685 and 580, the absorptive power shows decided maxima in the red, yellow, and green, that have no relation whatever to the emission. Dispersion.-H. Rubens (Wiedemann's "Annalen," October) finds that Helmholtz's electromagnetic theory of dispersion accords with results obtained for fluorspar, quartz, rock salt, sylvine, and one of the heavy Jena silicate flint glasses over the region of the spectrum that was investigated-54 octaves. Photometry.-J. B. Spurge (London Physical Society, Jan. 26) employs a photometric method in which two diffusing screens, illuminated respectively by the lights to be compared, are used as secondary sources, and the luminosities of the observed surfaces are adjusted to equality by varying the size of apertures through which the light passes from the screens to the surfaces. With small apertures this method admits of comparing different colored lights, since all colored lights appear as gray when the intensity is sufficiently feeble. E. S. Terry ("Physical Review," March-April) finds that, while the light transmitted by a rotating sectored disk equals the ratio of the total aperture to the entire disk, the effect on the retina is not in this ratio. With mixed light whose elements are of different luminosity the elements of low luminosity are intercepted most strongly. With a given light the error thus introduced increases as the aperture diminishes, but with ordinary illuminants it is negligible when the aperture is more than one half the disk. cbservations tell us nothing directly of the vibration of a luminous body, but only of the disturbance in the surrounding medium. The distribution of energy in the resulting spectrum may indicate a property of the medium rather than of the luminous matter. The vibration of the molecule, for instance, may be irregular, but the medium may take up and propagate some vibrations more quickly than others. 66 Spectrum Photography.-Edes and Valenta (Vienna Academy of Science) have obtained photographs of Bunsen-flame spectra by means of an apparatus that makes exposures of great length possible. A rotating disk of platinum gauze, mounted in a slanting position, dipped into a solution of the salt to be examined and passed at its upper edge through the flame. Prof. S. P. Langley, of the Smithsonian Institution (Paris Academy of Science, Aug. 13), has been able, by a perfected arrangement of the bolometer, and by automatically photographing the movements of the galvanometer needle, to furnish in one hour a complete and accurate record of the infrared region of the solar spectrum. The experiments of Victor Schumann on photography of rays of very short wave length are described in Naturwissenschaft Rundschau," No. 50. Very high ultraviolet photography has hitherto been interfered with by absorption in the prisms, lenses, and plate, and even in the atmosphere itself. Schumann, by discarding gelatin and using a pure silver-bromide plate, and by exhausting the camera, spectroscope, and spark-tube of air, has added to the known spectrum the region down to 100 micro-millimetres, and has exhaustively explored the hitherto doubtful region down to 185 micro-millimetres. G. Meyer (Wiedemann's" Annalen," March) has devised a new and ingenious method of photographing the spectrum of lightning. A diffraction grating ruled on glass is placed in front of the object glass, which is focused for an infinite distance. eral images of the flash are thus obtained, a central one from the undiffracted rays, and others corresponding to the different orders of diffraction spectra, the number of images of each order equaling the number of lines in the spectrum. Experiments with the apparatus seem to show a hitherto unobserved line in the lightning spectrum, at about wave-length 382 micro-millimetres. Sev Spectroscopy-The Rowland concave grating is astigmatic, which is an advantage as regards the elimination of dust lines from the spectrum, and the broadening of a star or spark spectrum into a band, but it has hitherto prevented the comparison of spectra by simultaneous observation. Dr. J. L. Lirks, in a pamphlet (Amsterdam, 1894), shows that the comparison may be made, provided the comparison prism or its equivalent, be placed at some distance from the slit, at a point determined by the intersection of the line joining slit and grating with a line drawn through the focus tangent to a circle whose center bisects the line joining focus and grating. Kayser and Runge (Berlin Academy, December, 1893) have made an exhaustive investigation of the spectra of the elements, discovering many new lines, including 14 for tin, 7 for antimony, and 22 for bismuth. The same (Wiedemann's" Annalen." May) have attempted to find uniformities in the structure. of the metallic line spectra, and find that the spectra of the metals investigated may be reconstructed by superposing several equal spectra differing by a constant oscillation frequency. Tin, lead, and arsenic require 3 such spectra, antimony 6, and bismuth 4. Janssen (Paris Academy of Sciences) has made new observations on the absorption spectrum of oxygen at high temperatures. The heating up to 300° C. was done by gas jets playing directly on the tube containing the oxygen, which was 30 feet long. Above 300° the gas was heated by raising to incandescence a platinum spiral by electricity. Temperatures of between 800 and 900° were thus obtained. The chief point brought out is the remarkable increase of transparency with increase of heat, the spectrum becoming brighter and extending its limits, especially at the red end. Thomas Ewan (London Royal Society, June 21) finds that the absorption spectra of very dilute solutions of several copper salts are identical. F. Paschen (Wiedemann's "Annalen," December, 1893) finds that there is no practical difference in the spectrum of a gas heated to 1,000° and the same burned in a Bunsen flame. He recognizes three types of light emission: (1) That of the vacuum tube-nearly all luminescence; (2) that of metallic vapors in Percy Frankland ("Chemical News," Jan. 19 the Bunsen flame-luminescence with true tem- and 26), in a review of the present state of knowlperature emission; (3) that of glowing gases-edge regarding the dependence of physical qualentirely due to temperature. Schuster ("Philosophical Magazine.” June), in a discussion of certain interference phenomena, points out that our Polarization.-K. E. F. Schmidt, in experiments on elliptical polarization by reflection ("Wiedemann's Annalen," May), finds that with glasses of equal refractive indices and different dispersive powers the glass with the higher dispersion shows the wider range of angle at which ellipticity is observed. G. Moreau ("Annales de Chimie et Physique," February) has investigated the magneto-rotary power of carbon bisulphide in the infrared part of the spectrum, and has succeeded in measuring it between 0-8 and 14 micro-millimetres, but the absorption of the bisulphide prevented observations at greater wave lengths. ities on atomic arrangement, lays stress on the experiments of Crum Brown and Guye, which show that rotary power is directly dependent on the combined atomic weights of the atoms constituting certain groups. 66 Phosphorescence.-Pictet (Paris Academy of Science, Sept. 24) has found that phosphorescent bodies cease to exhibit this property when exposed to very low temperatures. The phosphorescence is regained, however, when the bodies are allowed to assume their former temperature, though in darkness. He regards these facts as demonstrating that the phenomenon is due to molecular movement. H. Ebert (Wiedemann's Aunalen," September) has investigated the conditions of obtaining the best luminous effects by phosphorescence excited by electric radiation, and points out that the secondary circuit must be tuned to the primary, and that the condensers must have the least possible capacity. He has devised a lamp made of a glass globe containing a piece of phosphorescent material. Oscillations are conducted to tinfoil armatures on the globe, and produce vivid luminescence. The economy is very striking, but alternating currents of very high frequency are required, necessitating the employment of a transformer near the lamp, or perhaps in direct connection with it. Visibility.-P. L. Gray (London Physical Society. April 13) concludes from experiments: (1) That the minimum temperature of visibility is the same for a bright. polished surface as for one covered with lampblack, though the radiation may differ; (2) that the visible limit at the red end of the spectrum varies with the state of preparation of the eye, exposure to bright light diminishing sensitiveness and darkness increasing it; (3) that for the less sensitive condition the minimum temperature of visibility for the surface of a solid is about 470° C.; (4) that at night a surface at 410° C. is visible, and by resting the eyes in complete darkness one at 370° may be seen: (5) that observers' eyes differ somewhat in their minimum temperature of visibility. Standards.-Sharp and Turnbull ("Physical Review," July-August), as the result of a careful study of light standards, conclude that it is futile to attempt to obtain concordant photometric results from freely burning candles, though the English is more stable than the German. The most satisfactory results are given by standards that have chimneys to protect the flame. Electricity. Electrification. - Lork Kelvin and Magnus Maclean (London Royal Society, May 31) find that air can be electrified positively or negatively, but that it does not retain negative electrification so long as positive. For stable equilibrium it is necessary that the electric density, if not uniform throughout, diminish from the bounding surface inward. Hence a portion of non-electrified air must be wholly surrounded by electrified air. Prof. J. J. Thomson ("Nature," July 26) points out that these experiments prove simply that a gas can be electrified, and do not disprove the conclusion reached by him in his "Recent Researches in Electricity and Magnetism" that a molecule of gas can not be electrified, and that the charge of a gas is carried by separated atoms. J. J. Thomson ("Philosophical Magazine," April) confirms the idea already advanced by Lenard that the electrification developed by splashing drops is due, not to friction, but to a previous double coating of electricity on the drop, the water surface having one charge and the gas in contact with it an equal and opposite one. The splashing dislodges some of the external coating. This double coat, which is probably due to a tendency to chemical action, is possessed by the most diverse liquids, as water, turpentine, mercury, and molten metals. It may be on solids, too, in which case frictional electrification may be due largely to the rubbing off of the external electric coating. Conduction.-G. M. Minchin (London Physical Society, Nov. 24, 1893) has experimented on the action of electro-magnetic radiation on films containing metallic powders. Such a film in circuit with battery and galvanometer ordinarily acts as an insulator, but when the electrodes are brought very near together on the surface and one of them is touched with an electrified body a current passes. The electrodes may then be separated a little more and the process repeated till any desired extent of the film is rendered conducting. If the circuit be broken at the film, it becomes an insulator again, but if elsewhere it remains a conductor. A spark has an effect similar to that of an electrified body, and Prof. Minchin ascribes both to electric oscillations in the wires. Prof. Lodge, however, suggests that the phenomena are due to the increase of the range of molecular attraction, caused by electric polarization. The phenomena are akin to those of Minchin's impulsion cells (“ Annual Cyclopædia," 1890, p. 717, and 1891, p. 730) and of the conductivity of powders (1891, p. 731). Braun (“Zeitschrift für physikalische Chemie," February) has examined compound gases at the moment of their formation, and finds no conductivity when a Leyden battery is used with nitrogen dioxide and air: but when chlorine and hydrogen were tested with a Grove battery at the moment of explosion they were found to possess conductivity. When heated to 1,000° to 1.200° C., ammonium chloride and cadmium iodide conducted well, others only fairly or not at all. H. S. Carhart ("Physical Review," MarchApril) concludes that Sanford's experiments on the effect of the surrounding medium on the conductivity of a conductor ("Annual Cyclopædia," 1893, p. 621) are affected by some systematic error. Experiments of his own with more delicate methods than Prof. Sanford's show no such results as his. Bruno Piesch (Vienna "Berichte," May 25) finds that in acids and solutions of salts increase of pressure produces decrease of resistance, but the decrease is less as the pressure increases. Polarization in general increases with the pressure. C. V. Burton (London Physical Society, April 27) has proposed a theory of the mechanism of electrical conduction that is set forth in the following theorems, which he deduces from generally accepted theories of dielectrics and of magnetism: "I. In a region containing matter there may be (and probably always are) some parts which are perfect insulators and some parts which are perfect conductors, but there can be no parts whose conductivity is finite unless every finitely conductive portion is inclosed by a perfectly conductive envelope. II. In metals and in other non-electrolytes whose conductivity is finite the transmission of currents must be effected by the intermittent contact of perfectly conductive particles. III. If we suppose that in a substance at the absolute zero of temperature there is no relative motion among the molecules or among their appreciable parts, it follows that every substance at this temperature must have either infinite specific resistance (which does not imply infinite dielectric strength) or infinite conductivity." Convection.-Hurmuzescu (Société Française de Physique, reported in "Nature." Jan. 11) finds that if dissymmetry be caused between the 2 knobs of a Wimshurst machine by fixing a point to one of them, an electroscope placed at some distance becomes charged when the machine is worked. The charge is higher when the electroscope also has a point. That it is due to convection, not to induction, is shown by the fact that a metallic screen does not prevent it, whereas an insulating shade placed over the electroscope does. Discharge.-N. Piltchikoff (Paris Academy of Science) has devised a new method of studying electric discharges. One pole of a Voss machine is joined to a metal point, which is held over a copper dish connected to the other pole and containing a layer of castor oil. If the point is positively charged a depression is formed in the oil, and if a small screen is placed between the point and the oil there is produced in the center of the depression an elevation having the same size and shape as the shadow of the screen would have if the point were luminous. The phenomena are shown even when a powerful air blast is sent between the point and the oil. Electro-magnetic Radiation and Oscillation.Mascart ("Comptes Rendus," No. 6) has measured the velocity of propagation of Hertzian waves by calculating the period from the dimensions of the resonator and from experimental measurement of the wave length, using the data previously obtained by Blondlot, from which the latter thought he had shown a diminution of velocity with increase of wave length. A more accurate calculation of the frequency, made by Mascart, gives a remarkably constant velocity, whose mean is 303,200 kilometres per second. This is about 1 per cent. higher than the velocity of light; but Mascart thinks this due to the fact that the calculated value of the self-induction is too small. Birkeland and Sarasin ("Comptes Rendus," Nov. 6, 1893) find that when electric waves passing along a metal wire reach the end, that part of the tube of electric force nearest the wire turns about the end almost immediately; but the distant parts, not being capable of the same angular velocity, remain behind, and the tube therefore curves like a comet's tail. As the elements of the tube continue to move at right angles to their direction for the time being, the energy escapes from the end of the wire and is lost in the surrounding space. P. Lebedew (Wiedemann's "Annalen," August) has studied the mechanical effects of Hertzian waves on resonators at rest, and finds that when tuned to a higher pitch a suspended resonator, whether responding to the magnetic or to the electric components of the wave, is attracted; when tuned to a lower pitch it is repelled, the maximum effect occurring near perfect resonance. F. Sanford (~ Physical Review," July-August) has succeeded in photographing by electric vibrations by laying a coin on the sensitized side of a plate and connecting it with the terminal of a small induction coil, while a piece of tinfoil on the opposite side was connected with the other terminal. Prof. F. J. Smith by the same method, but using a higher potential, has obtained the same results in one second, while Prof. Sanford required from a half hour to an hour. Garbasso (Turin Academy of Science) finds that the electro-magnetic radiation reflected from a wooden plank, though of large wave length compared with the dimensions of the reflector, is not scattered, but reflected regularly. Zehnder (Wiedemann's Annalen," November) has polarized Hertzian rays elliptically and circularly by means of 2 wire gratings. Preece ("Engineering." London, Feb. 23) has made noteworthy experiments in inductive telegraphy. He has communicated across 3 miles of water by means of powerful alternating currents sent through a guttapercha cable 600 yards long on the mainland and 2 wires parallel to it on a distant island. He also telephoned by induction across Loch Ness, Scotland, 1 mile. H. Poincaré (Paris Academy of Science, Dec. 26, 1893) has shown mathematically that when an electrical disturbance moves along a wire, the head of the disturbance moves with such velocity that in front of the head the disturbance is nil, as in the case of light and of plane sound waves; but the electrical disturbance, unlike the others, leaves behind it a residue of finite magnitude. Kenelly ("Electrical Review," London, Dec. 15) finds that wires through which oscillatory discharges have passed are curiously bent sometimes at a right angle, and that some show microscopic holes or craters. These effects may be due to sudden expansion by heating. They are not influenced by a magnetic field. Platiniridium gives the best results. C. E. St. John ("American Journal of Science," October) finds that under very rapid electric oscillations the self-induction of iron wires is greater than copper, the difference varying from 34 to 43 per cent., and increasing as the diameter decreases. Prof. John Trowbridge, of Harvard ("Philosophical Magazine," August), finds that a unidirectional spark always excites an oscillatory discharge in a secondary circuit if its self-induction, capacity, and resistance permit. Thermoelectricity.-W. H. Steele, of Melbourne University ("Science." No. 562), in experiments on thermo currents in a single metal due to variations in temper, found that a sensitive galvanometer in circuit with a piece of iron wire showed a current when the wire was simply warmed with the fingers, but iron was the only metal that gave a current with a temperature below 100° C. A like effect was noticed in 12 different metals and 4 alloys. Gold gave the highest electro-motive force-05 volt; lead, copper, tin, zine, and antimony gave 03 volt; while the highest effect obtained from iron was 0.002 volt. The phenomena seem to be sufficient to mask ordinary thermoelectric effects at a red heat, hence the accepted thermoelectric tables are probably not trustworthy for high temperatures. G. F. Emery (London Royal Society, Feb. 8) has studied the thermoelectric properties of salt solutions, and finds that in a circuit formed by a metallic wire and a solution an |
