L -V ^ J 



the unextraded and intruded air, muft be in every ftroke added 

 to the reiidual air in the rec', as a conftant quantity ; which 



will 



Ttr.d to prevent tlie fault of tlie artill from being attributed to the flrufture of the 

 inrtrumcnt, I fliall alfo beg leave to add what follows: 



The quantity of air in meafure, that may in any length of time have forced an 

 entrance into the reC, may be found, and the juft allowance made for it in the de- 

 fe£i: of the pump's performance, by flating, as the Iicight of the merc>' in the ftand- 

 ard baroni', is to its afcent or defcent in the reduced or ftandard barom' gages, in 

 a given time, fuppofe fix hours, fo is the whole content of the rec and its ducts, 

 to the part thereof which would be filled in fix hours with the intruded air, if re- 

 duced to the denfity of atmofpheric air. Alfo, to know what allowance is to be 

 made for fuch intrufion in the defective altitude of the mere' in the gages, (late, 

 as fix hours, is to the time in which one ftroke is made (fuppofe half a minute) fo 

 is the afcent or defcent of the mere' in the reduced and ftandard barom^-gages, 

 in fix hours ; to the afcent and defcent which ought to be occafioned by the air 

 intruded, in the time of one ftroke ; then the afcent and defcent thus found, being 

 diminiftied or increafcd according to the ratio of the reC to the cyU, and deducted 

 from and added to the altitude of the mere' in thefe different gages, any remain, 

 ing difference between that altitude, and what exifts at llic time in the moft per- 

 fectly filled ftandard or reduced barom'', is to be attributed to the imperfedlion of 

 the machine. 



If the altitude of the barom'-gage, be referred to that of a ftandard-barom'' filled 

 with mere'' not boiled in the tube, though it were otherwife poured hot into it, 

 the tube itfelf being hot, and carfully purged of air by moving an air-buUble through 

 it the mere'' and tube being peifeftly clean, yet the altitude of the mere'' in 

 fuch a ftandard-barm'" muft be fuppofed to be lefs than it would be if the tube 

 were perfectly filled, by at leaft ,;|^ th parts of an inch ; becaufe as this air- 

 pump never failed in a favourable ftate of the air, to raife the mete'' in the gage 

 thus much higher than it ftood in a ftandard-baromr filled (feveral times) in the 

 above manner; it is plain this latter will be always imperfedlly filled and the 

 height of the mere' in it deficient, by the above quantity, and I found that it 



would 



