compact column. At times, twenty or thirty of these columns can be counted. A bluish steam accompanies the eruption, which rises straight into the air with great power, from many hundreds of fathoms.

In connection with this eruption, it is interesting to notice that during the nights of March 29 and 30 a heavy rain of ashes or sand took place along the west coast of Norway to the Swedish frontier, the whole country being covered with gray dust to such an extent that from a pint of snow more than a tablespoonful of residue was left after the snow had melted. The dust consisted of little irregular sharpedged grains, principally silicates, and probably originated from the eruption in Iceland.-12 A, XII., 75, and XI., 575.

DRYNESS OF THE SOIL IN INDIA,

In a memoir on the waterworks at Nagpur, Central India, Mr. Binnie gives a large amount of information with reference to the variability of the rainfall in India and other countries; and among the investigations into which he entered was one illustrating the dryness of the soil during the dry seasons, and the consequent amount of water absorbed by it after every rainfall. An area of 6 square miles, or 4224 acres, was drained by trenches into a reservoir, and the height of water in the latter subjected to careful observation. Three rain-gauges were also placed within this area, by means of which the rainfall could be determined. It was found that in the case of a measured fall of 2.24 inches of rain which fell in one hour and twenty minutes on the 18th of June, there was no perceptible drainage from this area into the reservoir, while on the 16th of September, in the case of a rain of 2.2 inches, which also fell in one hour and twenty minutes, the drainage into the reservoir amounted to over 33,000,000 cubic feet. These opposite results prove the extreme state of dryness of the soil in India at the end of the heated season, and its complete saturation after the heavy rains of the monsoon period. Of the drainage observed on the 16th of September, 98 per cent. entered the reservoir within two hours and fifty minutes. As to the question what percentage of total annual rainfall drains from the ground and can be emptied into reservoirs, Binnie states that almost every drainage area has, in this respect,

peculiarities of its own. Thus the average annual rainfall at Nagpur is 40.7 inches, of which 37.5 fall in the monsoon, and 3.2 in the dry season. Of this latter quantity, no part flows from the ground into the reservoir; and the records of the discharge of the drainage area are confined to about four months. The total depth evaporated during this time amounted to about four feet, or an average of about one fifth of an inch per day. The total loss of water during the season from the reservoir from all causes amounted to 104,000,000 cubic feet, out of which 55,800,000 cubic feet was evaporated, leaving 48,200,000 cubic feet as used or absorbed. An extensive comparison is given by the author between solar spots and India rainfall; but no very satisfactory conclusions are drawn by him.-Minutes of the Institution of Civil Engineers, XXXIX., Part I., 16.

MAGNETIC DISTURBANCES AND AURORAS IN THE ARCTIC

REGIONS.

Lieutenant Weyprecht reports that the magnetic disturbances in that portion of the arctic regions visited by the Austro-Hungarian North Polar Expedition are of extraordinary frequency and magnitude, and are closely connected with the aurora borealis; the disturbance being the greater and quicker the more convulsive the motion of the rays of the aurora and the more intense the prismatic colors. Quiet and regular arcs, without motion of light or radiation, exercised almost no influence upon the needles. In all disturbances the declination needle moved toward the east, and the horizontal intensity decreased, while the inclination increased. Movements in an opposite sense, which were very rare, can only be looked upon as movements of reaction. Weyprecht had expected to be able to connect the aurora with the galvanic earth current; but being far distant from the land, he was obliged to bury his connecting plates in the ice, in consequence of which the movements were too feeble to be observed. A similar failure attended the attempt to observe the atmospheric electricity. He found that storms followed almost every time after intense auroras, and thinks he is justified in the conclusion that the aurora is an atmospheric phenomenon, and closely connected with meteorological conditions.-12 A, XI., 368, 397.

A NEW SEISMOMETER.

A seismometer devised by Malvosia, of Bologna, is thus described: On a slightly inclined board is fixed a spherical cap, having eight grooves corresponding to the eight principal points of the compass. A little beyond the edge of the cap there is a projecting wooden ring which limits the inclined surface. On the top of the cap is poised a little brass ball slightly flattened at the point of contact. Upon the ball rests very lightly a eonical weight, by a small screw projecting from its base; which weight is suspended by a chain from an overhanging arm, movable up and down on a support at the side. The least shock will make the ball topple over. When it does so, it runs down one of the grooves of the cap to the inclined plane, at the lower part of which it finds a hole, and, passing into it, causes a gun to be fired off. After the ball leaves its position on the cap, a spring needle, longer than the diameter of the ball, shoots out from the little conical weight that rested on the ball, and catches in that groove of the cap down which the ball has run. Thus the direction is indicated from which the shock came. The instrument can be made very sensitive. It differs from that recently introduced by La Saulx into the earthquake stations of Prussia, principally in that the apparatus does not of itself record the exact time at which the shock took place, but merely calls the attention of the observer by means of the discharged gun. By combining the instrument of La Saulx with the simple upright pillar seismometer as described by Mallet, observations could be obtained which would be of more value than those given by the seismometer of Malvosia.-Journal of the Franklin Institute, April, 1875, 243.

THE MAGNETIC DECLINATION AT ST. PETERSBURG.

The long series of observations of magnetic declination that has been accumulating at St. Petersburg during the last 150 years has been subjected to a very full study by Mielberg, of that city, who has taken especial pains to examine the accuracy, reliability, and even genuineness of the observations, and to correct them for such sources of error as can be appreciated. The independent observations made

by Nervander at Helsingfors, from 1844 to 1848, afforded him an excellent point of comparison, and the means of determining the diurnal period of the declination. The annual variations could then be investigated, and finally the secular. From twenty-two years of observations, between 1841 and 1862, he has thus been able to compile a table representing the normal hourly variations of declination for each month, and for the entire year; from which it appears that the greatest western declination occurs between one and two o'clock A.M. in every month, being a little earlier in summer and a little later in winter, with a secondary maximum. just before two o'clock, and closely following the principal maximum. The greatest easterly declination occurs in the winter months between nine and ten A.M. During the rest of the year it varies between seven and nine A.M. The other maximum of easterly declination is not so decidedly expressed, and occurs in winter and spring between ten and eleven A.M. In respect to the secular variations of the declination, it would appear to have varied between two and a half degrees west in 1727, to six and a half degrees west in 1831, reaching its maximum about the year 1806, when it was about nine and a half degrees west. He suggests that if the increasing accuracy in our means of measurement is to have any meaning at all, and not be a useless expenditure of time and money, it becomes necessary now that, in any place where a magnetic observatory is surrounded by an assemblage of houses, there should be annually made, for purposes of comparison, an independent standard determination of the declination at some point near by but entirely outside of the buildings in question, and free from their influence.-Wild's Repertorium, IV., art. 1.

ATMOSPHERIC ELECTRICITY.

In reference to atmospheric electricity, Sir William Thomson states that if one ascend any mountain peak on a fine day, and there prove that the surface of the earth on the peak is negatively electrified, the result will be valuable to science; and if on several days the ground is found to be, all day and all night, negatively electrified, then there will be a very great acquisition to our knowledge regarding atmospheric electricity. According to him, positive atmos

E

pheric electrification of the air is merely inferential. What we know by direct observation is simply that the surface of the earth is negatively electrified; and many misleading and delusive statements in reference to the positive electricity of the air are to be found in encyclopædias and treatises on meteorology. Suppose, for a moment, that there were no electricity whatever in the air; that it was absolutely devoid of all electric manifestation, and that a charge of elec tricity were given to the whole earth-for which purpose no great amount would be necessary-such amounts as we deal with in our great submarine cables would, if given to the earth as a whole, produce a very considerable electrification of its surface. And suppose, in addition, which in fact seems to be shown by experiment, that all space above the atmosphere, and that the atmosphere itself were a non-conductor; then the charge could be given to the earth as a whole, just as a charge could be given to a pith ball electrified in the air of the room. Under these circumstances, all the phenomena that have thus far been brought to light by atmos pheric electrometers would be observed just as they are. The ordinary observation of atmospheric electricity would give just the result that has been obtained from it. The results that we obtain every clear day in ordinary observations on atmospheric electricity are precisely the same as if the earth were negatively electrified, and the air had no electricity in it whatever. Ordinarily we have evidence in the lower strata of the air of the presence of negative electricity; but in rainy weather it is sometimes positive and sometimes negative.-Journal of the Society of Telegraph Engineers, 1874, III., 12.

THE DESTRUCTIVE FLOODS IN SOUTHERN FRANCE.

The terrible floods which swept over several hundred miles of the northern portion of the Pyrenees and the plains at their base, appear, in the light of later and more reliable accounts, to have been far more destructive to life and property than was indicated by earlier reports. The Garonne and its affluents, which drain the larger portion of the Pyrenees, seem to have risen with such rapidity that at Toulouse, as an instance, a low-lying suburb, St. Cyprien, chiefly inhabited by the working classes, was overwhelmed almost