The Marquess of Salisbury. 323 



When using thermometers in the above experiments, they were 

 found to be very frequently pierced by the discharge at the sealed 

 point. This appeared to be due to the pointed form of the interior 

 space. A piece of glass tube was heated in the blowpipe and drawn 

 out, forming two closed tubes; the closed ends were then fused 

 together and slightly drawn out so that they were connected by a rod 

 of glass, the interior spaces of the tubes being pointed. When the 

 electrodes of the coil were inserted into the open ends of the tubes, 

 and the coil set in action, the intervening glass was pierced in less than 

 three minutes. A vacuum tube, with one end drawn out to a fine 

 point and sealed, was instantly pierced when the pointed end 

 was rested on the metallic plate connected to one pole of the 

 induction coil. Lord Salisbury observed that india rubber when 

 exposed to light undergoes a change, and a casual remark of his led to 

 the experiments described in a letter to "Nature," vol. 27, 1883, 

 p. 312. 



Arc lamps worked by primary batteries were used at Hatfield as 

 early as the winter of 1869-1870 to illuminate the south front of the 

 house on the occasion of the County Ball, and from 1879 to 1881 

 experiments were made on the lighting of the interior of the house by 

 electricity. At first the dynamos were placed in or near the house and 

 worked by a steam or gas engine. Afterwards the machines were 

 placed at the saw mills at a distance of more than a mile from the 

 house, where they were actuated by a water-wheel on the River Lea, 

 which runs through the park ; subsequently a turbine was used. 

 Hatfield House was one of the first large houses in the country to be 

 lighted by electricity. Electric motors were afterwards used in some 

 farming operations, especially for cutting weeds in the river. 



In 1881 and 1882 numerous experiments were made to obtain two 

 pieces of the same metal that would not produce an electric current, 

 when they were dipped in water and connected through a galvano- 

 meter. The experiments were first made with two sovereigns with 

 pieces of wet paper between them; on connecting the coins to a 

 galvanometer a current was always detected. Platinum wires and 

 rods were next tried, and notwithstanding their being cut from the 

 same piece of metal and being cleansed by action of many different 

 substances and by heating, it was not found possible to obtain them in 

 identical states. Two wires were heated by an electric current in an 

 exhausted tube, to which a side tube containing dilute sulphuric acid 

 was sealed, with the expectation of removing occluded gases ; but a 

 powerful current was produced when the tube was inverted so as to 

 connect the wires by the dilute acid. Very little regularity was 

 observed in the direction of these currents ; but at last it was found that 

 if a wire were left for some time in a solution of a permanganate, washed 



