fix a slender iron wire of three digits length through a round cork, so that the cork may support the iron in water. let this water be in a good-sized glass vase or bowl. pare the round cork little by little with a very sharp knife (so that it may remain round), until it will stay motionless one or two digits below the face of the water; and let the wire be evenly balanced. rub one end of the wire thus prepared on the boreal end of a loadstone and the other on the southern part of the stone (very skilfully, so that the cork may not be moved ever so little from its place) and again place it in the water; then the wire will dip with a circular motion on its own centre below the plane of the horizon, in proportion to the latitude of the region; and, even while dipping, will also show the point of variation (the true direction being perturbed). let the loadstone (that with which the iron is rubbed) be a strong one, such as is needed in all experiments on magnetick declination. when the iron, thus put into the water and prepared by means of the loadstone, has settled in the dip, the lower end remains at the point of variation on the arc of a great circle or magnetick meridian passing through the zenith or vertex, and the point of variation on the horizon, and the lowest point of the heavens, which they call the nadir. this fact is shown by placing a rather long magnetick versorium on one side a little way from the vase. this is a demonstration of a more absolute conformity of a magnetick body with the earth's body as regards unity; in it is made apparent, in a natural manner, the direction, with its variation, and the declination. but it must be understood that as it is a curious and difficult experiment, so it does not remain long in the middle of the water, but sinks at length to the bottom, when the cork has imbibed too much moisture.
