358 Mr. Nixon's Theory of the Spirit-Level. 



water within the wide one : and for the same reason the water 

 fills the horizontal tube (9), even when held above its general 

 level. No. 16 proves very clearly that a force of attraction 

 perpendicular to the axis of the tube does not elevate the fluid, 

 — or why should the included clastic air wholly prevent its 

 wonted ascent ? If such a force did exist, why should an al- 

 most insensible depression of the horizontal tube in No. 16, 

 overcome it so completely as to expel the bubble of air ? A 

 globule of air rises to the surface of a liquid in consequence 

 of the pressure upwards against its base being greater than 

 the downward pressure on its upper surface : hence the greater 

 the diameter of the globule, and the greater the force with 

 which it is urged upwards ; which may serve to account for 

 the tardy and impeded escape of the minute bubble on re- 

 peating the 16th experiment. Besides, as the small bubble has 

 to move at an almost insensible distance from the curved sur- 

 face of the glass, where the levity of the water must be very 

 great, its specific gravity approaches more to that of the water 

 in contact with its ends, and will therefore require a consi- 

 derable depression of the tube to make the vertical column 

 urging it below adequately heavier than the opposed one in- 

 cumbent on it. But when the bubble is long and deep, the 

 pressure upwards is derived from an incomparably longer and 

 specifically heavier vertical column acting against diminished 

 obtacles. 



(According to our theory, the specific gravity of the liquid 

 within the horizontal tube of a level will be least at the upper 

 corners of the tube ; but along the upper surface, and still more 

 so on the sides above the level of its axis, it will approach irt 

 a greater degree to its proper specific gravity. Hence the 

 surface of sections of the liquid, whether perpendicular to, or 

 in the direction of the axis, but especially ihejatter, will have 

 a concave curve at the extremities ; which will account for the 

 peculiar figure of ihe bubble of a spirit-level. 



(We may now comprehend how an uncurved tube contain- 

 ing a sufficient quantity of ether, &c. becomes possessed of a 

 bubble which does not extend to the ends of the tube. In a 

 level of this description, as its bubble, when the cylindrical 

 tube is horizontal, will be stationary indifferently in any part 

 of its length, and as the slightest mclination would displace 

 the bubble (if sufficiently large) indefinitely, causing it to move 

 at once to the elevated end of the tube, it could not serve to 

 measure the minutest variation of inclination.) 



Let the bubble of a curved tube pass over O'l inch for a 

 variation of inclination of 1"; then if its depth in the middle 

 be 0*2 inch, its length, when the tube is horizontal, should be 



equal 



