438 Life of Cavendish 



analysis, has lately treated the same subject in a very masterly manner. The 

 acknowledged imperfections, in some parts of Mr Cavendish's demonstrative 

 reasoning, have served to display the strength of a judgment and sagacity still 

 more admirable than the plodding labours of an automatical calculator. One 

 of the corollaries seems at first sight to lead to a mode of distinguishing positive 

 from negative electricity, which is not justified by experiment; but the fallacy 

 appears to be referable to the very comprehensive character of the author's 

 hypothesis, which requires some little modification to accommodate it to the 

 actual circumstances of the electric fluid, as it must be supposed to exist in 

 nature. 



4. A Report of the Committee appointed by the Royal Society, to consider of 

 a Method for securing the Powder Magazine at Purfleet. (Phil. Trans. 1773, 

 p. 42. Additional Letter, p. 66.) Mr Cavendish, and most of his colleagues on 

 the committee, recommended the adoption of pointed conductors. Mr Wilson 

 protested, and preferred blunt conductors ; but the committee persisted in their 

 opinion. Later experiments, however, have shown that the point in dispute 

 between them was of little moment. 



5. An Account of some Attempts to imitate the Effects of the Torpedo by 

 Electricity. (Phil. Trans. 1776, p. 196.) The peculiarity of these effects is 

 shown to depend in some measure on the proportional conducting powers of 

 the substances concerned, and on the quantity of electricity, as distinguished 

 from its intensity. Iron is found to conduct 400 million times as well as pure 

 water, and sea water 720 times as well; and the path chosen by the electric 

 fluid, depending on the nature of all the substances within its reach, an animal, 

 not immediately situated in the circuit, will often be affected on account of 

 the facility with which animal substances in general conduct the fluid. The 

 shock of a torpedo, producing a strong sensation, but incapable of being con- 

 veyed by a chain, was imitated by the effect of a weak charge of a very large 

 battery: and an artificial torpedo of wood being made a part of the circuit, 

 the shock diffused itself very perceptibly through the water in which it was 

 placed; but the experiment succeeded better when the instrument was made 

 of wet leather, which conducts rather better than wood, the battery being 

 more highly charged in proportion to the increase of conducting power. 



6. An Account of the Meteorological Instruments used at the Royal Society's 

 House. (Phil. Trans. 1776, p. 375.) Of the thermometers it is observed, that 

 they are adjusted by surrounding the tubes with wet cloths or with steam, 

 and barely immersing the bulbs in the water, since a variation of two or three 

 degrees will often occur if these precautions are neglected. For the correction 

 of the heights of barometers we have Lord Charles Cavendish's table of the 

 depression arising from capillary action. The variation-compass was found to 

 exhibit a deviation from the meridian 15' greater in the house of the Royal 

 Society than in an open garden in Marlborough Street; there was also a mean 

 error of about 7' in the indications of the dipping-needle, but it was difficult 

 to ascertain the dip without being liable to an irregularity, which often amounted 

 to twice as much. 



