82 Mr. Daniell on the Barometer, 



introduction of the metal. The conclusion he comes to is, that it is 

 always a portion of that which previously adhered to the glass, and 

 tfiat mercury is utterly incapable of absorbing either air or moisture. 

 One of his experiments is so simple, and at the same time so con- 

 clusive, that I cannot refrain from giving a short account of it. 

 He filled a barometer tube, and boiled it very carefully, and then 

 prepared a funnel made of a small capillary tube, which reached 

 through the mercury in the barometer-tube, to the closed end, and 

 was enlarged at the top. When introduced, it had been recently 

 made, and perfectly dry. Some mercury was then prepared by agi- 

 tating it in a bottle with water and air, and dried by means of fil- 

 tering paper, and afterwards passing it through paper cones, three 

 or four times, into dry vessels. A little of this mercury was poured 

 into the funnel-tube, and the air extracted by means of a fine 

 wire, so that the column was continuous. So much of this pre- 

 pared mercury was then poured in as fully to displace the mer- 

 cury which had been boiled in the tube. The barometer was 

 found to stand exactly at the same height as before in the same 

 circumstances ; and, when the mercury was heated, none of those 

 bubbles appeared which arose on the first boiling *. 



Still further to illustrate this subject, which I thought of the 

 highest importance, and to ascertain the difference of capillary 

 action in boiled and unboiled barometer-tubes, I undertook the 

 following experiments. The apparatus, which I made use of, con- 

 sisted of an upright pillar of brass, standing upon a mahogany 

 foot, upon which two horizontal arms of unequal lengths were 

 made to slide; at the extremity of each of these a steel needle, 

 with a fine point, was fixed perpendicularly downwards. These 

 points could be adjusted to the same plane, or their relative dis- 

 tance be measured, by means of a nut and screw upon the pillar, 

 which carried a nonius ; and the slightest contact of these points 

 with the clean surface of a basin of mercury was instantly per- 

 ceptible. I satisfied myself, by repeated trials, that the adjust- 

 ment might be depended upon to the one-thousandth part of an 



* Giornale de Fisica, Vol. vi. p. 20, or see Royal Institution Journal, Vol. 

 iv. p. 371 



