observed and illustrated. A bulb, three inches in diameter, is blown at the end of a glass tube eighteen inches long. In this a fine glass stem, with a sphere or disk of pith at each end, is suspended by means of a fibre of silk. The bulb is then perfectly exhausted and hermetically sealed. Instead of pith, disks may be made of iron, metal, cork, or other substances. The apparatus, when constructed with proper precautions, is so sensitive to heat that a touch of the finger on a part of the globe near one of the disks of pith will drive the index around over a quarter of a revolution, while it follows a piece of ice, as the needle follows the magnet. With a large bulb very well exhausted, a somewhat striking effect is produced. When a lighted candle is placed about two inches from the globe, the glass stem with its pith disks oscillates to and fro through gradually increasing arcs, until several complete revolutions are made, when the torsion of the suspended fibre offers a resistance to the revolutions, and the bar commences to turn in an opposite direction. This movement is kept up with great energy and regularity, like the movements of the balance wheel of a watch, as long as the candle burns. A modification of this apparatus, in which a glass thread is substituted for the silk fibre, allows quantitative as well as qualitative observations. The sensitiveness of the apparatus to heat rays appears to be greater than that of the ordinary thermo-electric multiplier. Thus the obscure heat rays from copper, at a temperature of 100°, after passing through glass, produce a deflection on the scale of 3 divisions, while under the same circumstances no current at all is detected in the thermo-pile.-Nature, XI., 494. ROOD'S APPLICATION OF ZÖLLNER'S HORIZONTAL PENDULUM.

The paper of Professor O. N. Rood, on the application of the horizontal pendulum to the measurement of minute changes and the diminution of solid bodies, although read in 1874, has only recently been published; and from it we learn the details of the instrument proposed by him as an improvement on Zöllner's horizontal pendulum. This proposed improvement consists essentially in an inflexible rod placed horizontally, and supported in that position in mid-air by vertical wires or springs, stretched in such a manner that the influence of gravity on the rod is no longer sensible,

while its motion is entirely under the control of the observer. Professor Zöllner's apparatus was designed expressly to measure attractive forces and slight changes of level. Professor Rood proposes to apply his own similar apparatus to the study of minute changes in the dimensions of solid bodies, for which purpose he gives it such dimensions as to impart to it an unprecedented delicacy. The main difficulty in the use of this apparatus is the fact that it is exceedingly sensitive to the most distant and unseen sources of disturbance. Thus Professor Rood remarks that children, playing at a distance of 360 feet, caused temporary deflections of one or two scale divisions; and similar deviations were caused by the lower notes of an organ in a neighboring church, the middle and higher notes producing no sensible results. These effects upon the apparatus can be eliminated, however, by making a sufficient number of observations, the evils caused by them being only temporary. As usual in all investigations, the effects of temperature are the most insidious.

As illustrating the fineness of the measurements that can be made with the horizontal pendulum, Professor Rood gives some figures showing that the one eighteen-millionth part of an inch becomes a sensible quantity; whereas hitherto, with the best optical and mechanical means, it has been hardly possible to measure the two one-hundred-thousandth part of an inch.-Am. Jour. Sci., 1875, IX., 441.

THE ELASTICITY OF BARS OF ICELAND SPAR.

Dr. G. Baumgarten mentions that a lecture of Professor Neumann on the theory of elasticity, in which he called attention to the interest that would attach to the determination of the co-efficients of elasticity in crystalline bodies, led him to undertake this labor, and that, so far as he knows, his own are, as yet, the first direct observations on the elastic properties of crystals. Voigt, however, has since then investigated the elasticity of the crystals of rock salt. Iceland spar was chosen by Dr. Baumgarten, among other reasons because, in reference to its physical and optical properties, it is better known than almost any other mineral. His determinations of its elasticity were made by measuring the bending of bars of spar, when pressed in various directions, and which had been cut in different directions from the crys

tal. The bars operated upon by him were two inches long, and had a square section of about the seventieth part of a square inch. He finds that the amount of deflection in the centre of a bent bar is a function of many quantities, but his observations allow him to state, first, that the deflection of a bar whose section is a perfect square is the same, no matter against which side the bending force is applied. Second, it varies with the dimensions of the bar as regards its thickness, breadth, and length, precisely as though the body were homogeneous; and the same laws apply to it within the limits of accuracy of his observations as apply to ordinary iron bars, the deflections being proportional to the cube, to the thickness, and to the length. Third, the deflection is dependent in a peculiar manner on the direction of the axis of the bar, in relation to the optical axis of the crystal from which it is cut. There exists, however, in this respect no symmetry with reference to the optical axis of the crystal. Bars cut parallel to the longest diagonal of the crystal give a minimum of deflection; those cut parallel to the shortest diagonal giving a maximum deflection.-Inaugural Diss., Berlin, 1875.

A NEW MANOMETER.

M. Fol has submitted to the Physical Society of Geneva a description of a manometer specially designed for deep-sea soundings. This instrument consists essentially of two spherical reservoirs, superposed, and connected by a capillary tube. The upper reservoir should be closed and filled entirely with a compressible liquid-for example, alcohol. The other sphere has an opening in its upper part, and is filled with mercury, which also fills the capillary tube. The quantity of mercury which shall have passed from the second reservoir into the first, when the apparatus has been submitted to a given pressure, will give the measure of this pressure, and consequently of the height of the column of water or the depth in the sea.-Mem. de Soc. d. Phys. de Genève, 1874, 483.

THE PHYSICAL PROPERTIES OF MATTER IN THE LIQUID AND GASEOUS STATES.

Professor Andrews, in a preliminary notice of his researches on the physical properties of matter in the liquid and gas

eous states, says that these investigations have occupied him continuously since 1869. In these he has experimented with gases under a pressure of 500 atmospheres. Of course, great difficulties have been experienced by him in measuring such pressures with accuracy; but the previous difficulties that he has experienced have been, or shortly will be, entirely overcome. His recent experiments fully confirm the conclusions published by him six years ago with reference to carbonic-acid gas, viz., that its contraction under great pressure is greater than it would be if the law of Boyle holds. strictly good. Under a pressure of 223 atmospheres, this gas is reduced to of its volume under one atmosphere, being slightly less than one half the volume it ought to occupy if it were a perfect gas, and contracted in accordance with Boyle's law. He infers, by analogy, that the critical points of the greater number of gases not hitherto liquefied are probably far below the lowest temperatures yet attained; and these substances are not likely to be seen, either as liquids or solids, until we can obtain much lower temperatures than those produced by liquid nitrous oxide. Again, the law of Gay-Lussac, like that of Boyle, is true only within certain limits and conditions of gaseous matter; in fact the co-efficient of expansion changes rapidly with the pressure, and if the pressure remains constant the co-efficient changes with the temperature. In reference to the law of Dalton, which is that the particles of one gas possess no repulsive nor contractive power with regard to the particles of another, Dr. Andrews's experiments show conclusively that this is not true; and that the so-called critical point is, for instance, lowered by the admixture of carbonic-acid gas with a non-condensible gas. The law also entirely fails when one of the gases is at a temperature not greatly above its critical point; it only holds good when these gases are at feeble pressures, and at temperatures greatly above their critical points.

ON THE INFLUENCE UPON THE MOVEMENT OF A PENDULUM OF A FLUID CONTAINED IN ITS SPHERICAL BOB.

The illustrious Bessel, in prosecuting his investigations into the force with which the earth attracts various bodies, employed a pendulum having a hollow cylinder of brass as

its bulb, in which he placed the various bodies to be experi mented upon. His observations gave him the result that the attraction of the earth was the same for all the bodies upon which he experimented; and his determination of the length of the simple seconds pendulum at Königsberg is one of the most correct we possess. He, however, found that when his cylinder was filled with water, the length of the seconds pendulum as computed for that substance was too great. The origin of this deviation Bessel attributed to the fact that the inclosed fluid was by the swinging of the pendulum set into vibrations of its own, whereby its moments of inertia in reference to the axis of vibration of the pendulum was different from what it would have been in the case of a uniform solid body. He accordingly found that the experiments made with long pendulums filled with water showed no such anomaly as in the case of shorter pendulums. Professor O. E. Meyer, well known for his investigations into the friction of gases and fluids, having suggested a somewhat different explanation, his student, Lubeck, has made this matter the subject of an inaugural dissertation, in which he considers the movement of the fluid contained within a pendulum, whose bulb is, for simplicity's sake, a hollow sphere instead of a hollow cylinder, Lubeck shows that the fluid contained in the hollow sphere is not set in motion by its rectilinear movement, but only by its oscillations about the diameter at right angles to the plane of the pendulum's vibration. This oscillation takes place with velocities which are constant for each spherical surface concentric with the hollow sphere, and any initial oscillatory motion is, in a certain time, destroyed by the inner friction, provided that it is originally of the same order as the veloc ity of the pendulum itself. After this time had elapsed, the motion of the pendulum is quite periodical. The extent of the arc through which the pendulum swings diminishes in a geometrical ratio when the time increases in an arithmetical ratio. The duration of the vibration of the pendulum is greater than if the same pendulum contained within itself a perfect fluid, instead of one having internal friction. The duration of the vibrations is smaller than if the fluid should be replaced by a perfectly solid body. When the length of the pendulum becomes very great the inner friction of the