52 PHILOSOPHICAL TRANSACTION'S. [aNNO 1724. 



to the liquor, which will always be right, both in the ball and the tube, when 

 the instrument is set upright. If by heat, the air at c be so expanded, as to 

 drive the liquor to the top of the tube, the cavity a will receive it, which will 

 come down again and settle at d, or near it, according to the level of the place 

 where the instrument is, as soon as the air at c returns to the same temperature. 

 To preserve the same degree of heat, when the different observations are made, 

 the machine is fixed in a tin vessel fe, filled with water up to gh, above the 

 ball ; and a very sensible thermometer has also its ball under water, that the 

 liquor at d may be observed in each experiment, when the thermometer stands 

 at the same height as before. The water is poured out when the instrument 

 is carried, which may conveniently be done by means of the wooden frame of 

 fig. 10, which is set upright by means of 3 screws, such as s, and a line and 

 plummet pp. The back part of the wooden frame is represented by fig. 11, 

 where, from the piece at top k, hangs the plummet p, over a brass point at n: 

 Mm are brackets to make the upright board kn continue at right angles with the 

 horizontal one at n. The 12th figure likewise represents the wooden frame 

 and screws. Figure 1 3 represents the machine seen in front, supposing the 

 forepart of the tin vessel transparent. And here the brass socket of the recurve 

 tube, into which the ball is screwed, has two wings at u, fixed to the bottom, 

 that the hall may not break the tube by its endeavour to emerge, when the 

 water is poured in as high as gh. 



As the tube is of a very small bore, if the liquor should rise into the ball a, 

 in carrying the instrument from one place to another, some of it would adhere 

 to the sides of the ball a, and on its descent in making the experiment, so 

 much might be left behind, that the liquor would not be high enough at d, to 

 show the difference of level ; therefore, to prevent that inconveniency, a blank 

 screw is contrived to shut up the hole at a, as soon as one experiment is made, 

 that in carrying, the air in a may balance that in c, so that the liquor shall not 

 run up and down the tube, whatever heat and cold may act upon the instru- 

 ment, in going from one place to another. 



Now, because one experiment being made in the morning, the water may be 

 so cold, that when a id experiment is made at noon, the water cannot be 

 brought to the same degree of cold that it had in the morning ; therefore in 

 making the first experiment, warm water must be mixed with the cold ; and 

 when the water has stood some time, before it comes to be as cold as it is likely 

 to be at the warmest part of that day, observe and set down the degree of the 

 thermometer at which the spirit stands ; and likewise the degree of the water in 

 the barometer at d ; then screw on the cap at a, pour out the water, and carry 

 the instrument to the place whose level you would know ; there pour in the- 



