*68 Account of a /elf-moving Lamp. 



•have to offer upon this fubjefl may be entitled to the notice of the Edinburgh Royal Scv 

 ciety, fo I will refer this point to your determination after you have had leifure to confider 

 the contents. 



The phenomena treated of in the feqael were quite new to me a few months ago, and, fo 

 iar as I know, have not hitherto been attended to or dcfcribed by any body elfe. What I 

 •have called the hydroftatical lamp, confifts of a fmall circular patch of common writing 

 .paper, about tliree-eighths of an inch in diameter, having about a quarter of an inch of foft 

 ^cotton thread {landing up through a puncture in the middle to ferve as a wick ; and the phe- 

 nomena inqueftion are certain motions which fuch minikin lamps acquire when lighted and 

 Tnade to fwim in very pure fallad oil. 



A (hallow glafs bafon, with fides rifing nearly perpendicular, or a common glafs falver, 

 will conveniently contain the oil for thefe experiment?. As foon as the lamp is lighted, it 

 will immediately fail brilkly forward in fome direction till it meets the fide of the vefiel, and 

 afterwards will take a circular co'urfe, always bearing up to the fides, and fo will perform 

 many revolutions. 



Sometimes the circulation is from right to left, and fometimes in the contrary dire£iion, 

 according as that point of the paper bafe, which in the diredt failing kept always foremoft, 

 ■turns away from the fide of the glafs, a little to the right or to the left hand of that which 

 comes to be the point of coiitadl. This turning away of what may be called the lead- 

 ing poiflt of the bafe is diftin6lly obfervable by a partial rotation of the lamp round the 

 wick as an axis, as foon as it arrives at the fide of the vefiel. Sometimes, though rarely, 

 the leading point itfclf attaches itfelf to the fide, and forms the vinculum, in confequence of 

 the well-known corpufcularattraftion between the elevation of oil around the bafe and that 

 belonging to the fides of the glafs; and when the vinculum fo correfponds to the leading 

 point, the lamp will be found to fland dill without any tendency to circulate. 



When the little witk has any fenfible eccentricity upon the circular paper bafe, the lamp 

 will fail fo as to make that part of the bafe which lies neareft to the wick the ftern ; and if 

 the bafe of the lamp be clipped to an oval form, and the wick placed in the longer axis ex- 

 centrical, that end of the bafe neareft the wick will alfo keep hindermoft, when the lamp 

 fails icrofs the falver. In the fame manner, if there be an equilateral triangle, having its 

 wick in the perpendicular which bifedls any of the fides, either the vertex or fide will be- 

 come the ftern, and keep hindermofl-, according as the wick is placed neareft the one or the 

 Other. Lamps fo conftrudled are found alfo to circulate upon their arrival at the fide of the 

 veflel, when the leading point turns away from the glafs, as it commonly happens. 



Whatever be the caufe of the failing of the lamp diredlty forward, the perpetual circu- 

 lation after it arrives at the fide feems to proceed from the force which formerly impelled 

 jt fiill ading in the fame manner, but in a direction inclined to that of the corpufcular at- 

 traction which forms the vinculum ; and it is evident that this inclination will be greater or 

 lefs according as the leading point is more or lefs averted from the glafs. When it fo hap- 

 pens that the leading point and vinculum coincide, it (hould feem that both forces juft now 

 mentioned muft urge the lamp in a direction perpendicular to the fide of the glafs ; in which 

 cafe it muft ftand ftill, agreeable to obfervation. 



The next thing which I had occafion to take notice of when the lamp failed in a direcfl 

 courfe, was a fcemingly very adlive repulfion between its ftern and the oil at the furface con- I 



tiguoas 



