of the tube to rise, pushing up with it its index, which is, in its turn, left in its place where the maximum thermometer has passed. The graduations of the second or minimum stem of the thermometer are counted downward, and those of the maximum stem are counted upward. As constructed by Casella, it is said this instrument is extremely sensitive, convenient, and reliable.-Quar. Jour. Meteor. Soc. of London, 1875, II., 193.

NEW SELF-RECORDING THERMOMETER.

In constructing a thermometer in which the dilatation of the metal shall give the measure of the temperature of the air, Tremeschini states that he has endeavored to eliminate the inconvenience peculiar to the nature of glass by making use of a metallic band as an indicator of the temperature. In his thermometer, which he exhibited lately to the French Meteorological Society, he employs a band of copper slightly platinized in order to preserve it from oxidation. This band is nine centimeters long and seven millimeters broad, and has a thickness of one twentieth of a millimeter, and is therefore extremely sensitive to atmospheric temperature changes: it is coiled about a central axis, very much like the hairspring of a watch, and is contained within a case similar to that of an ordinary aneroid barometer. The temperature is read on the face of the thermometer by an index, which may even describe an entire circle in passing from -40° to +100° Fahrenheit.-Nouv. Meteor., 1875, 14.

ON THE EXPANSION OF INDIA RUBBER BY HEAT.

According to the studies of Schmulewitsch, based in part on the studies of Puschl and Exner, as well as his own experiments, the somewhat anomalous behavior of caoutchouc under the influence of heat may be expressed by the following four propositions: First, caoutchouc is a body whose density is a minimum at a certain temperature. Second, this minimum temperature changes with the mechanical extension, being lower the more the body is extended by the application of some external force. Third, in the case of caoutchouc unexposed to any strain, the temperature of the minimum density is higher than ordinary temperatures, but approaches the latter by heating; its co-efficient of expan

sion is positive, but diminishes with increasing temperature. Fourth, in the case of strongly extended caoutchouc, the temperature at which its density is a minimum is lower than ordinary temperatures; its co-efficient of expansion is therefore negative at the latter temperature, and increases numerically with the temperature.-19 C, VIII., 146.

ON THE MOLECULAR HEATS OF SIMILAR COMPOUNDS. Professor F. W. Clarke states that as the result of an extensive comparison between the molecular heats of similar compounds, he finds that these have equal values, not at the same temperature, but at what are called corresponding temperatures, which are at equal or nearly equal distances from the respective melting points.-Bull. Phil. Soc. Washington, June, 1874.

ON THE REPULSION DUE TO HEAT.

In his reply to the criticism of Professor Reynolds, Professor Crookes states that abundant observations which have been accumulated by him during some years appear in every way to contradict the theory that the phenomena observed by him are due either to air-currents existing within vacuum tubes or to electrical phenomena. As to the theory of Professor Reynolds, that the effects are the results of evaporation and condensation, he satisfactorily shows that while this explanation might sometimes be admissible, yet in general it requires the adoption of assumptions that seem to be wholly at variance with the facts. He concludes by stating his belief that the repulsion observed by him as accompanying the radiation of heat and light is directly due to the impact of the waves upon the surface of the moving mass, and is not a secondary effect through the intervention of aircurrents, or electricity, condensation, etc. Whether the ethereal waves actually strike the object moved, or whether at the boundary of the surface, solid or gaseous, there are intermediate layers of condensed gas which, taking up the blow, pass it on to the layer beneath, are problems the solu tion of which must be left to further research; and, without insisting upon any theory of his own, he proposes it merely as a useful working hypothesis. Any theory will account for some facts, but only the true explanation will satisfy all

the conditions of the problem, and this can not be said of either of the theories which have thus far been discussed. To quote the eloquent language of Sir Humphrey Davy, "When I consider the variety of theories which may be formed on the basis of one or two facts, I am convinced it is the business of the true philosopher to avoid them altogether. It is more laborious to accumulate facts than to reason concerning them; but one good experiment is of more value than the ingenuity of a brain like Newton's."-7 A, XLVIII.,

94.

THE SPECIFIC HEAT AND CUBIC EXPANSION OF BODIES.

Mr. Walter Spring communicates to the Royal Academy of Belgium the following note with reference to the specific heat of bodies. He states that he sought to determine whether there were any relation between the specific heats and the co-efficients of cubic expansion by heat. He arrives at very beautiful results, both practically and theoretically. For instance, the computations which he makes of the specific heats of mercury and of graphite agree to the fourth decimal place with the observations of Regnault, Dulong, and Petit. He concludes that the product obtained by multiplying the specific heat of any body by its atomic weight can not be constant, since the specific heat is itself a function of a variable factor.-Bulletin of the Royal Academy of Belgium, 1874, 294.

LIGHTNING-CONDUCTORS.

In a few remarks upon the action of lightning-conductors, Secchi, the well-known astronomer, describes the storm of November, 1872, in which the cathedral and palace of Alatri were struck by lightning, these structures having been free from such visitations for many years. The damage done on this occasion was, as he shows, due in great measure to the fact that the lightning-rods, instead of being directly connected with the metallic gutters and other portions of the roof, were isolated from them. The fluid, therefore, sought to make its own way to such other good conductors as were After quoting other instances, he expressed the opinion that the conditions most favorable to safety consist in joining the lightning-rod directly to all the metallic portions

near.

of the roof, and especially to the rain-water pipes, in order that greater facility may be offered to the electric fluid in its passage to the earth.-American Engineer, I., 122.

ATMOSPHERIC ELECTRICITY IN SPITZBERGEN.

In reference to the observations of atmospheric electricity in high northern latitudes, in which, as yet, our instruments have generally given negative results, Wijkander states that the late Swedish polar expedition gave special attention to this subject, and that all their observations show that at relatively high temperatures the air conducts elec tricity very well, to which fact is ascribed the absence of lightning and the presence of the Northern Lights. It has been said that these latter phenomena depend upon the great moisture of the air in these regions; but it seems clear that the polar light is conditioned by other circumstances, since the same temperature and the same degree of humidity do not bring forth these results in other latitudes.--19 C, VII., 422.

VELOCITY OF THE TRANSMISSION OF ELECTRIC FORCE.

The question as to whether electric and magnetic forces. require sensible time to exert their influence, at a distance, has been made the subject of numerous investigations, one of the most interesting of which is that of Herwig, who has endeavored to conduct experiments upon as large a scale as possible. The preliminary results to which he was led have, he thinks, justified him in formulating the conclusion that if the terrestrial magnetic influence has any definite velocity of transmission whatever, it must be at least at the rate of half a million of miles per second; and that the influence of the earth's magnetism at any point of the earth's surface attains its full degree within of a second.-19 C, VIII., 30.

EDLUND'S THEORY OF THE NATURE OF ELECTRICITY,

In a report on the theory lately advanced by Professor Edlund as to the nature of electricity, Dr. Emsmann states that apparently Edlund has, in this matter, taken such a step forward as was made when previous investigators were able, by means of one æther, to explain both optical and thermal

phenomena. Edlund's theory consists essentially in ascribing to the æther itself an inertia which necessitates a slight interval of time in order to affect its movement. The flow of æther from one body to another explains the electro-dynamic phenomena, while its abundance or deficiency in any body serves to explain the electro-static phenomena. As regards the chemical influence of the galvanic current, it is assumed that the electricity has an equal influence upon the bodies that are to be separated or combined by it. The rotation of the plane of polarization of light is elucidated by the simple assumption that the electric æther is not dif ferent from the optical æther, and it must be acknowledged that Edlund's theory is based upon well-known facts, and is distinguished by its simplicity and sufficiency.-7 C, X., 402.

ON THE ELECTRICITY OF MINERAL WATERS,

Professor Theury, of Geneva, and Dr. Minnich have conducted some remarkable experiments in reference to the electrified condition of the mineral waters of certain springs, respecting one of which, the Stadthof, near Baden, in Switzerland, they state that their experiments show that the warm water at its escape from the soil is quite strongly electrified, it being negatively electrified with reference to the electric current at the thermal spring at Limmat. The currents observed by them are not the result of any thermo-electric action, nor are they the result of any special electric chemical action between the carbonic acid gas and the platinum electrode, but appear to them to be peculiar to the springwater itself.-13 B, III., 186.

THE STRATIFICATION OF ELECTRIC DISCHARGES IN VACUO.

Messrs. De la Rue, Miller, and Spottiswoode have conduct. ed a long series of investigations looking to the ascertainment of the cause of the stratification of electrical discharges in vacuo. Without bringing their investigations to a close, or pointing out any conclusions as distinctly reached, it is evident from their experiments that the stratification is due to a peculiarity in the flow of the electricity, which flow is apparent ly of the nature of an intermittent discharge, whose periodical overflows, so to speak, take place at very short intervals, and whenever the current acquires strength enough to over