ON AQUEOUS AND IGNEOUS METEORS. 559 



or connected with some well or drain : it should be of ample dimensions, 

 and where smallest, of copper, since copper conducts electricity more readily 

 than iron. In one instance a conductor of iron, four inches wide and half 

 an inch thick, appears to have heen made red hot by a stroke of lightning. 

 It seems to be of some advantage that a conductor should be pointed, but 

 the circumstance is of less consequence than has often been supposed.* 

 Mr. Wilson exhibited some experiments in which a point was struck at a 

 greater distance than a ball, and therefore argued against the employment 

 of pointed conductors. Mr. Nairne, f on the contrary, showed that a ball 

 is often struck in preference to a point. But it has been observed, that if a 

 point attracts the lightning from a greater distance, it must protect a greater 

 extent of building. It is easy to show, by hanging cotton or wool on a con- 

 ductor, that a point repels light electrical bodies, and that a pointed con- 

 ductor may, therefore, drive away some fleecy clouds ; but this effect is 

 principally derived from a current of air repelled by the point ; and such 

 a current could scarcely be supposed to have any perceptible effect on clouds 

 so distant as those which are concerned in thunder storms. In order to 

 escape personal danger in a thunder storm, the best precautions are, to avoid 

 eminences, and all exposed situations, as well as a near approach to conduc- 

 tors. The neighbourhood of windows, looking-glasses, fire-places, and trees, 

 must always be considered as hazardous. 



It has been supposed that a sudden condensation of the air, arising from 

 cold, accompanied by a deposition of moisture, and propagated by a con- 

 tinuation of the cause, by means of the cold occasioned by expansion, pro- 

 duces frequently the noise of thunder, without any lightning, and without 

 any electrical agitation : but it does not appear that the opinion is well 

 established. J 



The phenomena of waterspouts, if not of electrical origin, appear to have 

 some connexion with electrical causes. A waterspout generally consists of 

 large drops like a dense rain, much agitated, and descending or ascending 

 with a spiral motion, at the same time that the whole spout is carried along 

 horizontally, accompanied in general by a sound like that of the dashing 

 of waves. Spouts are sometimes, although rarely, observed on shore, but 

 generally in the neighbourhood of water. They are commonly largest 

 above ; sometimes two cones project, the one from a cloud, the other from 

 the sea below it, to meet each other, the junction being accompanied by a 

 flash of lightning : and when the whole spout has exhibited a luminous 

 appearance, it has perhaps served to conduct electricity slowly from the 

 clouds to the earth. Some of these circumstances may be explained by 

 considering the spout as a whirlwind, carrying up drops of water, which it 

 has separated from the surface of the waves ; and the remainder may per- 



* See Report of Committee appointed to consider of a Method for securing Pow- 

 der Magazines ; with Mr. B. Wilson's Dissent, Ph. Tr. 1773, Ixiii. 42. See also 

 ibid. liv. 247; Ixviii. 999. Cavallo, ibid. 1788, p. 1. Murray's Treatise on At- 

 mospheric Electr. 1828. Harris, On the Utility of fixing Lightning Conductors in 

 Ships, Plymouth, 1830. Annals of Electr. iv. 310 ; v. 41. 



t Ph?Tr. 1774, p. 79; 1778, p. 823. 



J See Harris's Essay on Thunderstorms, 1843. 



