As regards the monthly period we see that the ratio between the maximum and minimum is nearly twice as great at Armagh as at Sandwick; at Liverpool there is but little variation comparatively.

It may be worth notice, in passing, to point out that at none of the stations quoted is there any indication of a marked increase of velocity at the time of the equinox, either in spring or autumn, so that the idea of special equinoctial gales being recognisable in these islands is, as far as the above figures indicate. unfounded.

However, all discussions of wind are most laborious, for each individual hourly record must be resolved into its components, and the figures for individual years are so discordant that no period shorter than seven or eight years will give mean values worth having. Usually this resolution has been into two components, Northand-south, and East-and-west respectively, the opposite directions being regarded as positive and negative; but, as Professor Arthur von Oettingen and others have pointed out, such a method is insufficient, and the four components North, South, East, and West must all be given, each being regarded as positive, if we wish to discover the laws of annual and diurnal change of the wind. This is self-evident, because if during a given period, say a week, the wind has blown for six hours from West and for six hours from East, and if the velocities from the two directions have been the same, the resulting motion of the air, as far as these winds are concerned, is nil, while there has been a large amount of real movement, which is all masked if only two components are exhibited.

The truth of this statement is illustrated by the

following table, which shows the seasonal values of the components for the year 1873 at Dorpat.

referred to is mètres per second :—

The unit

The two columns of so-called 'resultants' give a very incomplete account of the actual movement of the air. In spring, for instance, the resultant motion from North is only 0.06, but when we look to the components we find that this value has been obtained by subtracting a motion from South of 1.02 mètres per second, from a motion from North of 1.08 mètres per second. All traces of this process are concealed if only the final value is given.

6

The whole problem of wind discussion is of a very complex nature, and its treatment is far from being settled as yet. On the whole, the best mode of publishing the records for different stations is the construction of wind-roses.' This is done by calculating the percentage proportion of the number of wind observations from each point of the compass and printing the results either in a tabular form or representing them by a diagram. Wind-roses may be made to show the force, as well as the direction, of the wind from the different points.

167

CHAPTER X.

ELECTRICAL PHENOMENA.

Atmospheric Electricity.--The subjects of hail and of the electricity of the atmosphere are closely connected, inasmuch as hailstorms are always associated with thunder and lightning. The direct proof of the identity of the electricity of a thunder-cloud with that obtainable from an electrical machine was furnished, about 130 years ago, by Franklin's famous experiment with a kite. Three years earlier he had suggested the employment of pointed insulated conductors to test the circumstances on a principle which will be explained at p. 183, but in June 1752, tired of waiting for the erection of such a conductor on a steeple in Philadelphia, he sent up a kite, attaching to the end of the string a key, and to the key a silken ribbon, which he tied to a post under a shed, in order to insulate the apparatus. As soon as the rain wetted the string, and thereby improved its conductivity, sparks were obtained from the key on presenting his hand to it. The experiment was immediately repeated elsewhere with better conducting lines, made of wire, and these yielded sparks or flashes of fire several feet in length. Such experiments are highly dangerous, owing to the very great tension of the electricity engaged, and within three months from Franklin's first experiment

a Russian physicist, Professor Richmann, who had arranged an apparatus to bring the electricity into his laboratory, was killed during a thunderstorm. He approached the end of the conducting wire, when a ball of fire apparently leaped to his head, and killed him on the spot.

It is beyond the scope of an elementary book to attempt a detailed description of the apparatus and methods employed in the measurement of atmospheric electricity, and it will suffice to quote the opening paragraphs of the section relating to the subject in the 'Instructions in the Use of Meteorological Instruments.'

'The simplest electroscopes, viz. the Gold Leaf, Bennett's, and Bohnenberger's, are sufficient to show the nature of the electricity present in the air, but it is always found that very little electricity can be observed near the ground, and in order to obtain satisfactory indications the conductor of the electroscope should be brought into contact with the air at some distance from the earth's surface, by means of a collector.

A simple rough method of collection is to shoot a metallic arrow upwards into the air, the arrow being tied to one end of a conducting string, the lower end of which carries a ring which rests upon the electroscope. The arrow being shot upwards, the electroscope will be found to be electrified as it mounts; and when the ring leaves the plate, the instrument will indicate the state of electrification of the air at that point where the arrow is at the time.

"This manner of observing electricity is simplified by substituting a long conductor reaching upwards. A gilded fishing-rod may be employed, its lower extremity being insulated.

The usual method employed, however, is Volta's, in which the electricity is collected by means of a flame burning at a height, either in a lantern hung to an insulated mast, and connected to the electroscope by a wire, or by a slow-burning match, attached to the top of a long metal rod. The electricity of the air, in the neighbourhood of the flame, by its inductive action upon the conductor, causes electricity of the opposite nature to accumulate at the upper extremity, where it is constantly carried off by the convection currents in the flame, leaving the conductor charged with electricity of the same kind, and potential, as the air.

"The principle of Volta's method has been made use of by Sir W. Thomson in his water-dropping collector, now employed in observatories, and found to be extremely useful for the observation of atmospheric electricity.

'A copper can is placed on an insulating support, which may be of ebonite, having the surface thinly coated with paraffin; or of glass surrounded with pumice-stone soaked in sulphuric acid. From the can a small pipe projects a considerable distance into the air, and terminates in a fine orifice. The can being filled with water, and the tap which opens into the jet pipe turned on, a small stream of water is allowed to flow out, care being taken that it is so small that it shall break into drops immediately after leaving the nozzle of the tube.

In half a minute from the starting of the stream, the can will be found to be electrified to the same potential as the air at the point of the tube.

This collector cannot be employed during the time of frost, unless means are adopted to prevent the freez