Fig. 2. 



same 



of air, a, 



140 Frederick Guthrie on a Sensitive Mercury Barometer. 



face of the mercury in the capillary tube would then move at 

 a rate proportional' to the sectional areas of the vacuum end 

 and the capillary. I believe this plan was not found to answer, 

 on account of the oxidation of the free end of the mercury. 



The plan which I submit, and which in the models I have 

 made and have had made appears to succeed well, more 

 resembles that of M. Duppa than of any other with which I 

 am acquainted. One form, shown in fig. 2, made for me by 

 Messrs. Cetti, of Brooke Street, Hol- 

 born, consists of an ordinary baro- 

 meter-tube, B, 6 millims. in internal 

 diameter, connected by a flat hori- 

 zontal spiral, S, of 2 millims. internal 

 diameter, with the open tube C of the 

 diameter as B. In S is a bubble 

 at such a distance from B 

 that it cannot enter B when the mer- 

 cury in that tube is at the top. The 

 motion of the bubble is, of course, 9 

 times as fast as that of the level of 

 the mercury in either limb, or 4^ that 

 of the true barometric variation. The 

 object of bringing B andC together is 

 to avoid as much as possible the effect 

 of the relative differences of height 

 on change of place. In another 

 form (which is being made by Mr. 

 Hicks, of Hatton Garden), the limb 

 C is made to enclose hermetically the 

 limb B. This should bring the effect 

 of such relative tilting to a minimum. 

 In one which T have myself made, the tube S being straight 

 and placed on a stone mantlepiece, a drop of sulphuric acid is 

 employed instead of the air-bubble. It has the effect of making 

 the motion rather more nimble, but is perhaps not quite so 

 trustworthy for exact measurements, because the wetting of 

 the tube by the liquid causes some inconstancy in its calibre. 

 A tube S of 2 millims. internal diameter, connected with a 

 barometer-tube having at its upper mercurial surface a dia- 

 meter of 20 millims., would exaggerate barometric motion 100 

 times ; and of course the sensibility would approach the nearer 

 to twice this, the greater the diameter of the open limb. There 

 is indeed absolutely no theoretical limit to the possible sensi- 

 bility. It must be left to practical glass-workers to determine 

 whether a spiral form is the best for the tube S, or whether a 

 series of parallel tubes joined alternately end to end are to be 



