104 



PROCEEDINGS OF THE AMERICAN ACADEMY 



TABLE IV. — CoMPAKisoN of Tait Gauges Nos. 1 and 4. 



No. 1, L = 100 cm.; Aq ~ .27 cm. \ A^ = \ cm. 

 No. 4, L = 58 cm.; Aq = .27 cm. ; A^ = 1 cm. 



In series of the kind given in Table III., the agreement of observa- 

 tions occupying similar positions in similar series is always very close, 

 the mean difference being only 1.2 atmospheres. Nevertheless, the 

 data of Table III. still show decided hysteresis. (See Fig. 6.) The 

 maximum horizontal distance apart of the " on " and " off" series is 

 about 40 atm. Comparing Tables I., II., and III., therefore, the per- 

 sistence of the apparent lag phenomenon may therefore be considered 

 proved. I further substantiated this inference by grading the cycles. 



19. Errors involved. — It is expedient to examine the chief errors 

 involved. The discrepancy in question cannot be due to an imperfect 

 Bourdon mechanism ; for in adjusting the pressure at any given value 

 just before observing, the actual pressures must as frequently be in- 

 cremented as decremented, both in the "on" and the off" series. 

 Again, the liquid adhering to the walls of the capillary of the Tait 



