308 Self-Registeri7ig and Indcstructable Barometer. [April 



meter 15 pounds ; it should be about two inches above the centre of 



gravity. 



The above remarks were set down as memoranda cn first vpaAhng 

 over the papers to which they refer ; bu ill 

 reading, an objection has offered, wh Ue 

 balance barometer ; — this I will now 

 proceed to explain — Let aa repre- 

 sent the surface of the quicksilver, 

 which being large with regard to 

 the tube b b, we may suppose to be 

 fixed, since we can make allowance 

 for its fluctuations ; c c represents 

 the surface of the quicksilver, cor- 

 responding to any given pressure, 

 (29 inches for instance) — Now the 

 weight imposed upon the barome- 

 ter-end of the balance, is that of the 

 tube, and of a cylinder of quicksilver which may be imagined to sur- 

 round the column d d, which is supported by the pressure of the 

 atmosphere. Let us suppose the atmospheric pressure to be increased 

 to 29, 5 inches ; what now takes place ? The surface c c rises to e e, 

 and consequently an additional weight (proportional to e e) is imposed 

 upon the barometer-end of the balance; — now the effect of this weight 

 being to lower the whole tube bodily, it follows that a further rise of 

 the surface e e will take place, and that, successively, cause and effect 

 will ensue, until the entire tube is filled, or the end of the tube " brought 

 up" by the bottom of the cistern. 



The form No. 2 seems to claim attention principally from being con* 

 structed of a metallic tube. Leaving this consideration out of the 

 question, it is by no means clear that the plan proposed is more free 

 from objections than many which at present exist. How are we 

 assured that a bubble of air does not exist in the quicksilver between 

 e and c ? For, allowing the stop-cock to be air-tight, still, in the 

 course of carrying, the quicksilver in the short leg being in contact 

 with air, and being agitated forward and back, will assuredly some- 

 times take up a bubble of air, and thereby render the reading erro- 

 neous. 



With regard to the self-registering apparatus, it appears to be well 

 adapted to the purpose intended ; but the objections to the baro- 

 meter, which it is proposed to be employed with, must be surmounted, 

 when the self-registering barometer will no doubt be brought into 

 general use„ 



