1835.]^ Filling Barometers. . 443 



A very slight inclination is sufficient for this purpose, and 

 of course, any increase thereof tends to diminish the lowest 

 bubbles by compression : but a little change from less to 

 greater, and vice versa, puts in motion the stationary ones, 

 vehen there are such. If the tube is not clean, little 

 bubbles will fix themselves to any dusty part, and cannot 

 sometimes be moved unless by washing them away : pour- 

 ing the mercury gently from the head of the tube to the 

 finger, and back again three or four times. 



When the mercury lies smooth for its whole length, the 

 finger is to be withdrawn, the hair put in and the 3 or 4 

 inches vacant carefully re-filled. The tube is then to be 

 stopped and inverted in the quicksilver with the same pre- 

 cautions as before, against the entry of a bubble under the 

 finger. When brought to the perpendicular position, it 

 should be turned round and examined on all sides, to see 

 that the column is perfectly smooth and bright ; and when 

 quickly inclined, so as to allow the mercury to reach the 

 head, it should return a smart rap. If both these conditions 

 are found, the tube is well filled ; but as a repetition of the 

 levelling and vibratory process for drawing off" the bubbles, 

 is a work but of little time and trouble, it is better per- 

 formed for the sake of security. 



The tube thus twice purified from air, and replaced in the 

 mortar of quicksilver, wants only its cistern (6). The lid is 

 first to be plunged beneath the surface, and there slid up 

 over the tube, say three or four inches, where it is to be 

 fixed by a slight wedge, or slip of paper. The box is next 

 to be filled, plunged also under the quicksilver, and its 

 edge passed under the tube, which must rest in it, not quite 

 upright, so that it may be full to the head. The box and 

 tube, in this position, both full of mercury, are to be taken 

 out of the mortar and set on a saucer or plate ; when the 

 tube being brought upright, mercury will descend, and flow 

 over the sides of the box ; more is also to be withdrawn from 

 the box, by suction with the pouring tube, until about \ inch 

 deep is left above the bottom of the tube : a little more or 

 less, according to the state of the barometer, at the time, 

 above or below the average ; but if below, the average can 

 be attained by inclining the tube. 



A slip of wood, say | inch square, but cut away at each 



