ON THE SUPPOSED CURRENTS IN HOT LIQUIDS. $5 



Solution below. Yet, during the firft part of the cooling, the- 

 amber was moving upwards and downwards; particles of it 

 often come to the very furface, and defcended again to the 

 bottom. This experiment proves decitivcly that the motion 

 of the amber f§ not occasioned by currents, and that there 

 were, in reality, no currents at all during the whole time of, 

 cooling from 1.50° to 50°. 



Having thus proved that the motion of the amber is not Why does die 

 owing to currents, but that it forces its way through the i^J"^ m6v? • 

 liquid; it becomes a point of fome confequence to af certain,^,;/ f 

 if pofhble, the caufe of this motion. Perhaps the following 

 obfervations will contribute to throw fome light on the fub- . 3r ,j ,. 

 jccl. 



1. When an alkaline folution containing amber, as exactly Obfervation i. 

 as poffible of the fame fpecific gravity with itklf, is heated ^^"g^ 1 

 in a glafs veilel nearly to the boiling temperature, and then heat than the al- 

 fufpended in the air to cool, the whole of the amber finks to kaline folution. 

 the bottom. Hunce we fee that, though the amber is of the 



fame fpecific gravity with the cold folution, it is* fpeciftcally 

 heavier than the fame folution when hot. Confequently 

 amber does not expand fo much when heated as the alkaline 

 Jolution does. 



2. But the amber does not remain long at the bottom ; the Obs. a. Each 

 grains of it gradually rife one after another, and afcend to the f s a ^ r ^ „"" 

 furface, but with very different velocities, iome rapidly, others an adherent air- 

 flowly. After touching the furface they always fall »3own.^^^^ 

 again to the bottom. '. Thefe alternate motions continue for f aC e, and then 

 fome time ; but the number of afcending and defcending par- falls *5* ,n » 

 tides gradually diminifhes, and at laft they all fettle at the^ - :: 

 bottom. It is no uncommon thing for two particles of amber, 



one afcending, the other defcending, to meet together in 

 fome part of the folution. In that cafe they fometimes ex- 

 change motions ; the one that was formerly afcending now 

 finking to the bottom, while the defcending particle returns 

 again to the furface. If the experiment be made with a 

 tranfparent folution, and in a good light, every afcending 

 particle of amber will be obferved to have an air bubble at- 

 tached to it which buoys it up. This bubble feparates at the 

 furface of the liquid, and the amber, thus deprived of its 

 buoy, tumbles down again to the bottom. When two par- 

 ticles meet, the air bubble fometimes panes from the one to 



