idly. Gradually the gas was forced out, and the lower part of the bag doubled up into the upper part, forming a sort of parachute. He landed just twelve miles from his startingplace, having evidently retraced a portion of his track.

ATMOSPHERIC ELECTRICITY.

An important work of Edlund on the Origin of Atmospheric Electricity is published by the Swedish Academy, and also in the London, Dublin, and Edinburgh Philosophical Magazine, and in Hann's Zeitschrift. He shows that the rotating magnetic earth, being a good electric conductor, must, by unipolar induction, give rise to a nearly uniform charge of negative electricity throughout its own whole mass, but to a variable charge of positive electricity throughout the atmosphere. The discharges, of course, give rise to aurora and lightning. He deduces, with comparative accuracy and simplicity, the diurnal and annual periods, and the geographical distribution of these phenomena.

Important papers by Angot in the Annuaire of the Meteorological Society of France, and by Everett in the report of the Permanent Committee of the Vienna Congress, present the best connected accounts of modern methods and theories that are at present accessible to ordinary readers.

In a note on the Origin of Thunder-storms, Professor Tait explains how a pair of vertical rotating columns revolving in opposite directions can be produced out of one column in the upper regions of the atmosphere revolving about a horizontal axis. He also suggests that the source of the electricity specially developed in thunder-storms may probably be found in the contact of air with the surface of the warm drops of

water.

A general review of the subject of Atmospheric Electricity is given by Dr. Margules in the Vienna Zeitschrift, wherein he expounds the electrical principles that occur, and enumerates some of the questions that first demand investigation.

The phenomena of Globular Lightning are described by M. Fitzgerald, of Donegal County, Ireland, who saw a globe of fire in the air descend gradually along the crown of a ridge, and down into the valley, where it drifted along a boggy surface, occasionally disappearing in the soil, but reappearing farther on. It finally flew across the stream, and buried it

self and disappeared in the peat bank. Its total duration was about 20 minutes. It appeared about 2 feet in diameter at first, but gradually diminished to 3 inches. The sky was clear at the time. Wherever it touched the ground in its course, it ploughed up the earth to a depth of several feet. In the discussion on this report, Mr. Symons stated that forked lightning and globular lightning were, he thought, the same.

The Scientific Gazette for Venezuela publishes a communication to the Physical Society of Caraccas, by J. M. Tebar, on an Electric Phenomenon peculiar to Lake Zulia, called the Farol de Maracaibo. He explains this light as being an electric phenomenon due to a quiet discharge of a large extent of atmosphere of the negative electricity with which ascending currents of air are saturated, while the descending currents are charged with positive electricity. The same journal contains an original theory, accompanied by some mathematical demonstrations, relative to the Nebular Hypothesis, by E. Ricard.

Hildebrandsson has published an Investigation into the Thunder-storms of Sweden, based on observations at about 250 stations, from 1871 to 1875. He distinguishes between the thunder-storms that attend the advancing sides of extensive storms (the Wirbelgewitter) and those that originate in overheated districts (the Wärmegewitter). Similar classifications have been made by Mohn in Norway and Fron in France, and are occasionally alluded to in the weather reviews of the Army Signal-Office. Scarcely a single instance was recorded in the five years in which it could not be shown that the so-called "heat" ΟΙ sheet lightning was simply the reflection of lightning so far distant that the thunder was inaudible, or possibly refracted above the observer's ear.

66

Dr. Pissis publishes in his "Physical Geography of the Republic of Chili" some notes on thunder-storms in that state. Destructive thunder-storms are unknown in the inhabited portions of the state, but are of daily occurrence among the mountains. In the morning a small cloud forms around each summit, which soon increases to a large cumulus, and eventually extends over the whole range of peaks. In the afternoon fearful thunder and lightning occur; and, high up the mountain-sides, snow and hail. The hail and snow are phosphorescent, as it were, with electricity. The storms on the

eastern slopes of the Andes and on the pampas are of shorter duration than on the west side. The lower horizontal surface of the clouds, which is at an altitude of 1500 or 2000 meters, sometimes extends from the coast eastward to far beyond the Andes. The greatest thunder-storms in Chili last three or four days, but the electric discharges all occur in the first hour.

Carl Weyprecht has published the "Magnetic and Aurora Observations of the Austro-Hungarian Arctic Expedition of 1872 to 1874," with which he also gives an excellent analysis and résumé of our knowledge on the subject. Among other things, he strongly inclines to the opinion that the auroral light belongs to the lower portion of the atmosphere. He fails to establish any connection between the aurora and the weather. He concludes that the zone of maximum frequency moves northward during the winter, and again southward during the spring.

Thalen has a memoir on Exploitation for Iron Ore by Means of the Magnetic Needle. He gives full and strictly accurate methods, and his formulæ may possibly be applicable to the search for the seat of the disturbances that accompany auroras.

An elaborate paper on the Protection of Buildings from Lightning, by Dr. Mann, is republished, with additional notes, in the sixth volume of the "Professional Papers on Indian Engineering.'

The American Academy of Sciences of Boston has instituted an inquiry with reference to the general phenomena of lightning, and, among other things, has requested the Chief Signal Officer of the Army to collect, as far as possible, all notes and observations upon accidents caused by lightning. All who may read this note are cordially invited to record such facts as may come to their knowledge.

The Proper Method of Protecting Buildings by Lightningrods is discussed, in a fifth note, by Melsens, in the Brussels Bulletin, 1878, p. 43. He gives an analysis of the conditions necessary to insure efficacy, and discusses the constructions published by the French Commission.

According to Mr. Symons, both Mishel, Secretary of the French Commission on Lightning-rods, and Jarriant, a large dealer in lightning-rods adopt as the area protected from

lightning by a given rod a circle whose radius is 1 times the height of the rod above the building.

An excellent paper on Lightning-conductors was read before the British Association, by Mr. R. Anderson. He states that so slow has been the march of progress since the days of Benjamin Franklin, that lightning-conductors are still wanting on at least half, and perhaps two thirds, of all the public buildings, and on 95 per cent, of all the private houses in Great Britain. The terrible losses occasioned by lightning are due to three sources of neglect: first, not providing any lightning-rods at all; second, not placing them in the right position; third, not having them regularly tested, so as to ascertain their constant efficiency. Lightning-conductors, he says, ought to be tested at least once a year. Between three and four thousand pounds sterling were spent in protecting the Houses of Parliament some twenty years ago. Since that time they have never been tested, and there is no guarantee whatever that a discharge of lightning may not at any time fall upon the Queen's throne. The testing should take place at regular intervals.

The wonderful sensitiveness of Bell's telephone-such that it responds to induced currents of the strength of only the one thousand-millionth part of a C. G. S. unit, or less-gives it great importance as an instrument of research in relation to atmospheric electricity and terrestrial ground-currents. Some observations made in Providence, R. I., have led to the conclusion that by it the existence of a thunder-storm may be detected when otherwise altogether invisible at the station.

The Edison microphone has even been applied successfully to the observation of subterranean sounds produced by volcanic actions, and may evidently be further applied to whatever goes on in the earth and ocean.

Mr. Henry Goldmark, of the Laboratory of Harvard College, contributes observations upon the Effect of Temperature on Atmospheric Electricity. He used Sir William Thomson's water-dropping apparatus and his quadrant electrometer. The observations were made in a room whose temperature could be varied as desired. He found, first, that even a very considerable change of temperature does not have any great or marked effect upon the electric potential

of the air; second, that, however, a rise in temperature does produce a slight but constant increase in the potential.

OPTICAL PHENOMENA.

Montigny gives a preliminary chapter of results of his observations on the Scintillations of the Stars. The first interesting observation that he publishes is in the Bulletin of the Royal Academy at Brussels, 1878, p. 157. It was first noticed by Usher, 1788, at Dublin, that the stars scintillate remarkably during and preceding an aurora. In 1840 Arago says that Forbes and Neckar and himself agreed that the stars never scintillate unless there is an aurora somewhere. Montigny observed an especial increase of scintillation of auroras April 5, 1870, and June 1, 1878. He also says that the scintillation is connected with a lowering of temperature, and that this cooling occurs simultaneously with auroras and scintillation, and that, finally, the cooling causes the scintillation. In the second communication, in the Bulletin, Brussels, vol. xlv., p. 391, he gives the results of Seven Years' Observations with his Scintillometer on Fifteen Stars as to Color, and finds that the red colors predominate during dry weather, while the blue precedes rain. The memoir on the Connection between Scintillation and Rain, etc. (Bulletin, 1878, p. 598), is received too late to allow of the extended notice that it deserves.

A memoir by Wild (St. Petersburg Bulletin, vol. xxi., p. 312), on the Photometric Determination of the Diffuse Light of the Sky, gives the preliminary results of an investigation that for many years has engaged his attention. The instrument that he has invented and used for this purpose he designates as the urano-photometer, and it seems to combine the features of the photometers invented by Arago, Hirn, Wild, and Zöllner. A disk of ground glass illumined by the sun affords an artificial standard light, whose whiteness is turned to the required shade of blue by receiving it through a quartz plate and polarizing apparatus. Wild finds (1) the color of the diffuse sunlight, as we proceed from the sun northward on a vertical circle, changes gradually from the red to the violet end of the spectrum, and at a distance of 80° from the sun is nearly between Fraunhofer's lines C and D, or at the wavelength 0.000628 meter; from here onward to the horizon the