If we place a globe in a vessel of water, so that the vertex shall only just be covered, and place the globe eccentrically in the vessel so that the centre of the vessel may not be too far from the outside of the globe, and then impart an equable but slow motion to the water, in the manner of a vortex; by viewing the reflected light of the sky from the surface of the water above the globe, we shall be able to trace a succession of dimples, originating at I and O, and passing off with the current, and dying away. The direction of the fluid in these little eddies, will be the same as the direction of the current in the main vortex. If we displace the globe, so as to remove it far from the centre of the vessel, and impart the same motion, the vortex I will be found at E, and the direction of the current will be contrary to the direction of the fluid in the vessel. In the case of the earth and moon, the displacement can never change the position of the inner vortex much. It will always be to the right hand of the central vortex in north latitudes, and in consequence of the ether striking our globe in such a position, the current that is deflected from its true path, by the protuberance of the earth forcing it inside, is prevented by the circular current of the parts nearer the axis of the vortex, from passing off; so that a vortex is formed, and is more violent, _ceteris paribus_, than the vortex at O.
Whether this mode of action has been correctly inferred, matters little; the lateral vortices follow the law of such a position. The inner vortex always precedes the central from five to eight days, when ascending in this latitude, and comes to the meridian after the moon. The outer vortex, on the contrary, follows the central in its monthly round, and comes to the meridian before the moon. It will be readily understood that if the axes of these lateral vortices be produced through the earth, they will pass through similar vortices in the opposite hemisphere; but as the greatest latitude of the one, corresponds to the least latitude of the other, the same calculation will not answer for both. The same remark applies to the central vortex also.
Thus there are six passages each month over latitude 41°; but as there are intervals of 3° to 6° between two consecutive passages of the same vortex, it may happen that an observer in the middle latitude, would perhaps see nothing of their effects without looking for them. Generally speaking, they are not only seen, but felt. The time of the passage of the outer vortex ascending, corresponds so nearly (in 38° of latitude) at certain times, with the passage of the central vortex descending, that the two may be considered one if attention is not directed to it. The orbits of these lateral vortices depend, like that of the central vortex, on the orbit of the moon for eccentricity, but the longitudes of the perigee will not correspond with the longitude of the moon's perigee. This follows from the theory. As the elements of these orbits are only approximately determined, we shall confine our calculations to the orbit of the central vortex.