402 Mr. C. Barus's Comparison of the Bourdon, 



Table II. — Cyclic comparisons of the Bourdon Gauge 

 (Fraunhofer Micrometer) with the Amagat Manometre. 



Bourdon. 



Amagat. 



Bourdon. 

 cm./40. 



Amagat. 

 atm. 



Bourdon. 



Amagat. 



Bourdon. 



Amagat. 



em. '40. 



atm. 



cni./40. 



i 

 atm. 



cm./40. 



atm. 



*19-09 



* 



6-12 



914 



1-18-92 



t 1 



6-33 



901 



Kr04 



228 



7-94 



753 



i 15-83 



242 



8-20 



756 



1612 



227 



8 36 



753 



1 15-87 



239 



8-39 



739 



1193 



523 



11-20 



535 



11-91 



523 I 



11-08 



539 



1219 



518 



11-28 



534 



1220 



513 



11-08 



539 



8-40 



772 



1518 



251 



8-86 



730 



14-80 



270 



1 8-58 



765 



15-18 



254 



9-18 



720 



14-80 



274 



5-23 



969 



18-97 







5-25 



969 



19-06 







5-66 



956 







570 



956 







6-02 



914 







616 



935 







First series. 



t Second series. 



4. It appears from Table I. that if allowance be made for 

 errors of reading, the observations of the first two series lie 

 on a straight line. The factors, Bourdon/ Amagat, in the first 

 series are *998 in the on inarch and *987 in the off march. In 

 the second series these factors have changed to '984 and *992 

 respectively. It follows from both series that within 500 

 atmospheres a cyclic march does not appreciably occur. 



In the third series, which shows a change of rate above 

 700 atmospheres, on, the return inarch is distinctly cyclic, 

 and at 500 atmospheres the zone shows a breadth of almost 

 50 atmospheres. The zero is regained within 10 atmospheres. 

 The mean factors are '991 below and 1*008 above 700 

 atmospheres, on. The factor of the off inarch is of no interest. 

 Remarks of the same kind apply in cases of the fourth series, 

 where the factors are *982 below and 1*072 above 700 atmo- 

 spheres, on. Some irregularity in this series is due to un- 

 warrantably vigorous tapping. 



Thus it appears that the inference drawn in the earlier 

 paper* from a comparison of the Bourdon and the Tait 

 gauges is valid. The breadth of the cycles here was about 

 40 atmospheres. It is gratifying to note that even the 

 standard atmosphere used was correct to 1 per cent. The 

 present results, however, go further : since the whole of the 

 cyclic motion is in the Bourdon gauge, the indications of the 

 Tait gauge will probably be found very near the truth. 



5. My reasons for regarding the mechanism of the gauge 

 as relatively perfect were given elsewhere (/. c.) ; but it is 

 necessary to verify this conclusion by a direct-reading screw- 

 micrometer. Care was taken to fix the axis of the screw 



* Phil. Mag. [5] xxx. p. 346. 



