the Land's End about 4 hours. The tide-wave then runs along the south coast of England, and the north coast of France, to the Straits of Dover, which it reaches about 11 hours after the moon's transit. It runs also along the west coast of Ireland and Scotland, and reaches the Orkneys about 9 hours after the moon's transit. From thence it enters the German Ocean, and runs along the east coast of Britain, so as to reach Peterhead about 12 hours after the moon's transit, and Harwich in about 12 hours more, where it meets the tide-wave which had come through the Straits of Dover derived from the same Atlantic wave about 12 hours earlier. The tides of the German Ocean are produced by the mixture of these two tide-waves, and hence follow complicated laws as for the same reason do those of the Irish Channel.

The tide-wave which brings the tides to the eastern coast of North America appears to reach the southern parts about 7 hours, and the northern parts of the United States about 11 hours, after the moon's transit; but its course has not yet been distinctly traced.

How the tides on the eastern and on the western shores of the Atlantic are connected has not yet been clearly shown. It is difficult to explain the tides of the Atlantic Islands (Madeira, Teneriffe, &c.) by any simple form of a tide-wave.

It is remarkable that the European tide-wave, though following the moon's transit at a definite interval (nearly), moves (at first) in a direction opposite to the moon-namely, from west to east.

If we go to the Pacific, we find the same phenomenon. The tides on the western shore of South America, near Cape Horn, also move from west to east. They are simultaneous with the moon's transit at Chiloe; 1 hour after at Cape Pillar; and at Cape Horn it is 3 hours later than this.

Along a large portion of the west coast of the Pacific, it seems difficult to say whether the tide-wave travels northward or southward. From the Isthmus of Panamá, however, it appears plainly to travel to the northward, occupying about 12 hours to run from Realejo to Nootka Sound.

In the western parts of the Pacific the tide-wave runs to the westward, as we learn by its progress along the coasts of New Zealand and Australia, where the movement is better known than on any coasts out of Europe. It visits New Zealand about 6 hours, and Australia about 10 hours, after the moon's transit at Greenwich.

In the central parts of the Pacific the tides are small and anomalous (for they do not clearly depend on the moon), and hence it is still more difficult to connect the littoral tides than in the Atlantic Ocean.

The outer regions of the Pacific, broken by large islands, and the Indian Ocean, have tides, of which the laws of progress are more complex, and have not yet been disentangled.

The Diurnal Inequality (23, &c.) adds to the complexity of the tides. This inequality appears very conspicuously in the tides on the west

coasts of Europe and the east coasts of North America; but its maximum in those two regions does not appear to be simultaneous. It is very large in the Indian Ocean and on the coast of Australia, having different phenomena at different places, as noted in 26 and 27.

The movement of the tide along the surface of the ocean may be in some measure represented in the following manner: Draw lines through all the places where it is high water at the same time; that is, one line (generally it will be a curved line) through all the places where it is high water at One o'clock; another line through all the places where it is high water at Two o'clock; and so on. These lines, being the lines at which the tide is contemporaneous, are called cotidal lines. They represent the form of the tide-wave which carries the tide from one point of the shore to another.

Such cotidal lines have been drawn (in the Phil. Trans. for 1833 and 1836, by Dr. Whewell) for those shores on which the tides are best known, and especially for the coasts of Europe.

But it appears that we cannot, by means of such cotidal lines, express the movement of the tides in oceanic spaces. The cotidal lines can only be drawn in the neighbourhood of coasts.-(Phil. Trans., 1848, Part I.)

The best way to disentangle the phenomena of the tides when we are observing them at any place is to refer the time of high water and low water to the time of moon's transit; and to do this at once, while the series of observations is going on. For want of following this rule, it has very often happened that long series of tide observations have been made, which could not be turned to any use afterwards; and in almost every case the usefulness of such observations is by this method much increased, and the labour much diminished.

ARTICLE IV.

SECOND DIVISION, SECTION 1.

TERRESTRIAL MAGNETISM.

BY MAJOR-GENERAL EDWARD SABINE.

[The Article on Terrestrial Magnetism has been re-written for the present Edition. The description and use of the Unifilar Magnetometer in Appendix No. 1, and an addition to Appendix No. 3, have been contributed by Mr. Welsh, of the Kew Observatory.-E. S.]

1. THE magnetic observations which have been made and are at present making by naval officers have for their object the determination of the amount and direction of the Earth's magnetic force in different parts of the globe.

2. The amount of the magnetic force at any point of the Earth's surface may either be measured in absolute value, or its ratio may be ascertained to the value of the force at another station where its absolute measure is already known. No means have yet been devised for measuring absolute values at sea; consequently, all determinations of the magnetic force on board ships are necessarily of the relative class; these give the ratio, or proportion, which the force at the geographical position in which the ship is at the time when an observation is made, bears to its value at some land station which is included in the same series of relative observations, but where an absolute determination has also been made. Ships are therefore supplied with instruments for both absolute and relative determinations; the latter to be used chiefly at sea, but also on land at times when the ship is in harbour; the former to be used exclusively on land.

Absolute Measurement of the Magnetic Force.

3. No satisfactory method has yet been generally practised for the direct absolute measurement in one operation of the whole magnetic force of the Earth (called the "total force") at any particular point of its surface. But that portion of the force which acts in a direction parallel to the surface of the Earth (called the "horizontal component") may be measured with considerable accuracy by a process, of which the following brief description may suffice to give a general idea. If a magnet be suspended horizontally by a few fibres of silk, and made to vibrate in the horizontal plane on either side of its position of rest, the square of the number of vibrations in a given time is a measure of the horizontal component of the magnetic force of the Earth. But this measure is dependent on the individual properties of the magnet employed; and these properties influence the time of vibration; first, by the greater or less magnetic force which the magnet itself possesses; and, second, by the effects of the form and weight of the magnet, The latter effect, that of the form and weight of the magnet, may be eliminated when the moment of its inertia is learnt ; and this may either be calculated by known rules, or may be ascertained experimentally by vibrating the magnet 1° in its usual state, and 2° with its moment of inertia increased by a known amount. The influence of the magnetic force possessed by the magnet may also be eliminated by ascertaining its magnetic moment. This is accomplished by using it to deflect a second magnet similarly suspended in another apparatus. The deflecting magnet is placed at one or more well-measured distances from the centre of the suspended magnet, and perpendicular to it. The defleotions thus produced (i.e. the angular differences in the positions of rest of the suspended magnet, 1° when influenced solely by the Earth's magnetism, and 2° when in equilibrium between the Earth's magnetism and that of the deflecting magnet at the distances employed) furnish the ratio of the forces exerted respectively by the Earth's force

and that of the magnet; and as the product of the same two forces is given by the vibrations of the deflecting magnet when suspended as in the experiments first described, the values of either force may be separately ascertained. The influence of the magnetism of the magnet, and of its form and weight, being thus eliminated, a measure is finally obtained of the force of the Earth's magnetism, independent of the individual properties of the magnet employed in the determination.

4. The numerical expression by which the measure of the Earth's force thus obtained is denoted, depends on the units of time, of space, and of mass employed in the measurements and calculation. In conformity with the Instructions published under the authority of the Royal Society, a second of time, a foot of space, and a grain of mass, are the units so employed. The horizontal component of the Earth's magnetic force has been found, by the observations hitherto made, to vary at different points of the Earth's surface from 0 to about 84 of the scale founded on the units which have been specified.

5. Wherever the horizontal component of the force has been ascertained in absolute measure, there also, if the magnetic direction be known, the "total force” in absolute measure is determined; since it consists of the horizontal component multiplied by the secant of the angle which the magnetic direction makes with the horizon. As ships are supplied with instruments by which this angle, called the dip or inclination of the needle, is measured, the observations on land, when the ship is in harbour, give determinations of the total force, which serve as base determinations, to which are referred the relative results obtained at sea in the passage from one station of well-assured absolute determination to another; a practice corresponding to that which prevails in determinations of longitude, where stations of wellassured longitude are taken as base stations, to which intermediate observations are referred. The total force of the Earth's magnetism, expressed in the scale in which the British units already referred to are employed, has been