342 



Mr. G. Gore on the Thermo-electric [Feb. 23, 



form shown in fig. 7, each tube being filled with a conducting liquid, and 

 its outer end closed by a cork, in which was fixed a second platinum wire 

 to dip into the liquid. 



Fig. 8 represents the apparatus ; A A is a wooden stand supporting a tin 

 box, B. The box is water-tight, and has in its lower surface a long semi- 

 circular cavity (shown by dotted lines) to receive the upper ends of the 

 twelve tubes. To the back of the box is fixed a short cylinder of tin, C, 

 closed at its outer end. When the apparatus is in action, the box is filled 

 with hot water, and the water kept boiling by means of a lamp placed be- 

 neath the tube C. The twelve tubes were kept in position by divisions of 

 wood fixed to the back of the stand, as shown in the figure. 



The tubes 1, 3, 5, 7, 9, and 11 were filled with a previously boiled and 

 cooled mixture of J of an ounce of sulphuric acid, and 19 ounces of di- 

 stilled water; and the others, viz. 2, 4, 6, 8, 10, and 12, with a similarly 

 prepared solution of 110 grains of hydrate of potassium dissolved in 19 

 ounces of distilled water. 



The platinum wires were connected, in the order shown in the sketch, by 

 means of small binding-screws not represented in the figure. 



On connecting the terminals with a galvanometer containing about 180 

 turns of moderately coarse copper wire, and applying heat to the upper 

 electrodes and ends of tubes by means of the boiling water, no deflection of 

 the needles took place ; but on substituting a Thomson's reflecting galvano- 

 meter, which offered a resistance of 3040*7 B.A. units (=77872-327 miles 

 of copper wire ^ of an inch thick), a deflection of 40 degrees was readily ob- 

 tained, the hot platinum wire in the dilute acid being negative, and that in 

 the alkali positive, as shown by the direction of the arrows in the sketch. 



From these results it is evident the quantity of the electric current pro- 

 duced was exceedingly small, and its intensity considerable. By employing 

 electrodes of larger surface, such as spirals of platinum wire and more con- 

 centrated liquids, the quantity of the current would be very largely in- 

 creased. (See Phil. Mag. 1857, vol. xiii. p. I.) 



Fig. 9. 



Fig. 9 represents a simpler arrangement of this apparatus, in which only 



