(_b_) The conical armature within the receiver was removed, and the stream allowed to connect with the centre of the pump-plate. When the magnet was excited, the stream was violently projected at right angles to the poles, with a vibrating movement to either side according to the direction of the current. When the wire was positive the movement was towards the left, with a slight inclination towards the N. pole. When the wire was negative the movement was to the right, but in a rather strong curve towards the N. pole. The vibrating motion was very distinct, and gave the appearance of six or seven streams running off at regular intervals (see Plate XVIII. fig. 12). [Sidenote: Rotating wire over S. pole.] (_c_) The wire was next placed over the centre of the disk forming the S. pole. With the wire negative and the pole positive, rotation of the stream was decidedly, but not very strongly, from right to left from the centre of the plate (as the hands of a watch). With the wire positive and the pole negative, rotation was strongly left to right, with a disposition to spiral twist in the stream (see Plate XVIII. fig. 13). [Sidenote: Same over N. pole.] (_d_) The wire was placed over the disk forming the N. pole. With the wire negative and the pole positive, rotation of the stream was left to right. With the wire positive and the pole negative, rotation was right to left (see Plate XVIII. fig. 14). [Sidenote: Stream thrown across the receiver above the magnet-poles.] (_e_) The stream was thrown across the receiver from the lateral binding-screws above, and at right angles to, the disks, and afterwards in the opposite direction, _i. e._ along them. In neither case was there any marked change when the magnet was excited. (_f_) The conical armatures were placed with the pointed ends upon the disks in the receiver, and the stream thrown above and along them. It diverged—one part running straight across between the electrodes, whilst another stream and some cobwebs ran from each electrode to its nearest pole. The streams and cobwebs flickered a good deal. There was no marked change when the magnet was excited. _Some of Baron Reichenbach’s Magnetic Researches tested._ [Sidenote: Baron Reichenbach’s researches.] In 1846 Dr. W. Gregory published an abstract of Baron Reichenbach’s ‘Researches on Magnetism and on certain allied subjects, including a supposed new Imponderable.’ [Sidenote: Auroræ considered to be magnetic lights. Flames seen by “sensitive” persons.]