228 On the Gold Fish of Franklin. 



With the same charge, and scraps tolcrahly alike in shape, 

 the long ones hover at greater distances than the short ones. 

 This is most plainly seen when the experimenter succeeds in 

 causing several scraps to swim at the same time. 



When the charge diminishes in intensity, the leaves slowly 

 approach the knob of the jar; when, however, by means of the 

 knob of a second jar, electricity is communicated to the former, 

 the leaves again recede. In this manner, when the air is dry, a 

 scrap of gold-leaf can be kept swimming for hours. 



In some cases the rotation first begins when the floating leaf 

 comes near to the jar, and the rotation is accelerated as the leaf 

 approaches. 



With scraps of a suitable shape, the equilibrium is so stable 

 that the upright jar may be suddenly pulled downwards, with- 

 out altering the hovering condition of the scrap of gold-leaf. 



From the side of the leaf most distant from the jar electricity 

 is given off of the same name as that wherewith the jar is 

 charged; and it has the power of propagating itself to a consi- 

 derable distance. When a gold-leaf electroscope is placed near 

 the " fish," a permanent divergence is observed after the instru- 

 ment has been removed. 



When the swimming leaf is removed from the neighbourhood 

 of the knob, by suddenly withdrawing the jar, it is found that 

 those pieces which swim at about an inch distance are in a neu- 

 tral condition ; while those which swim at a less distance are 

 negative. Only such as, while swimming, come very near the 

 knob of the jar, show sometimes positive electricity. 



In a note to this paper Prof. Poggendorff remarks: — "Al- 

 though the more recent treatises on electricity do not mention 

 the fact above described, still it is not new, the experiment 

 having been made by Franklin in 1749 on the conductor of an 

 electric machine (see Experimental Observations on Electricity, 

 &c. London, 1769, p. 72). The free hovering of a piece of 

 gold-leaf, in so stable a position, must always be regarded as 

 a most remarkable fact, which is deserving of revival, particu- 

 larly as very few appear to be acquainted with it/' 



In the following volume of the Annalen, p. 164, Prof. Riess 

 makes the following remarks : — " A bit of gold-leaf, differently 

 pointed at its ends, and with its blunter end towards the con- 

 ductor of an electric machine, flies towards the latter, and re- 

 mains floating at a distance from it, because the electric wind 

 generated at the blunter end drives it back, while that generated 

 at the sharper end and the electric attraction drive it towards 

 the conductor. A point on the side of the strip of gold-leaf 

 must cause an axial rotation. The same action may be observed 

 with a sewing-needle which has its eye broken off. When it is 



