260 Mr. V. Guthrie on a Relation between 



lost all power of discharging the + electricity of the electroscope 

 it still retains its power, only slightly diminished, of discharging 

 *- electricity. This experiment is perhaps the most striking 

 one of this immediate series. The temperature at which the 

 difference of behaviour of the iron towards the two electricities 

 is most noticeable is a dull red heat, called, I believe, a cherry- 

 red. Attempts are made in the sequel to approximately measure 

 this crucial temperature. 



§ 19. Experiment. — The wliite-hct ball is removed from the 

 fire on the insulating hook, and its power of discharging -f- and 

 — electricities frequently compared as it cools. As with the 

 earth-connected ball of § 18, its power of discharging — en- 

 dures long after its power of discharging + is lost. 



§ 20. Experiment. — At a dull red heat, a ball, whether insu- 

 lated or earth-connected, will not discharge a positively charged 

 electroscope at a distance of half an inch. 



§ 21. Experiment. — At a dull red heat a ball will discharge 

 *— electricity when earth-connected, but not when insulated. If 

 the insulated ball which has, by cooling, just lost its power of 

 discharging — electricity be held half an inch above the electro- 

 scope and momentarily touched with the finger or a wire, the 

 leaves instantly but only partially collapse. At each touch a 

 further collapse takes place. 



§ 22. The whole aspect of the experiments with hot iron balls 

 is of a nature to convince the actual experimenter that in all the 

 above cases the collapse of the leaves is due to the escape of 

 their prevalent electricity, and not to the accession of electricity 

 of the opposite kind; for in no case do the leaves diverge after 

 collapsing. A good many experiments of various kinds, which 

 are sufficiently self-suggestive to release me from a description 

 of them, showed that the leaves of a neutral electroscope were 

 in no instance induced to diverge by the approach of hot bodies. 



§ 23. We have so much greater control over the heat of an 

 imperfectly conducting wire than over that of a hot metallic ball 

 (we can retain the former constant, both in regard to its tempe- 

 rature and position, so much better than the latter), that I was 

 soon induced to substitute the former for the latter. But as we 

 know that galvanic currents passing along a wire exercise a 

 locally wide range of galvanic influence, and that such influence 

 is especially manifested when the current is started or stopped, 

 approached or withdrawn from the influenced body, or increased 

 or diminished in strength, it became necessary to examine whether 

 under any conditions the current affected the electroscope when 

 little heat was developed in its passage. 



§ 24. Eaperiment. — The current from eight platin-zinc cells 

 was passed along a copper wire 1 inch thick and 9 inches long, 



