of H, F. Regnault, 



279 



closely with the experimental results up to a compara- 

 tively high temperature. 



Log. F = a — 6 771* — c n*, 

 where x = temperature in C. degrees + 20°, F = tension 

 in millimetres, a = 6.2640348, log. h = 0.1397743, log. c. = 

 0.6924351, log. m = i.994049292, log. n = l.998v43862. 



The values given by this formula for all temperatures 

 between — 20° and + 230° do not differ from the cor- 

 responding values derived from experiment by more than 

 l-400th of the whole, a difference, which, as Regnault 

 remarks, falls within the limits of observational errors. 



I have compared the results given in this memoir with 

 those obtained from experiments, made in the physical 

 laboratory of Union College, during February 1877, by 

 myself and others, under the supervision of Mr. 0. H. 

 Landreth, now of Dudley Observatory. The values are 

 in close agreement up to tensions of about two and a half 

 atmospheres; after this the correspondence is less exact, 

 though still quite marked. A comparison of the curves 

 constructed from the two sets of observations shows that 

 Regnault's curve may be regarded as the mean to which 

 the other approximates more or less closely, the sinuosi- 

 ties of the latter curve lying about equally on both sides 

 of the former.* It is probable, therefore, that if addi- 

 tional observations could be had, the curves would be 



' Comparison between some of the results obtained in the laboratory at Union 

 College and Series t of Regnault. 



Regnault. U. C. Laboratory. 



Temperature. Tension in mm. Temperature. Tension in mm. 



106°.81 963.16 107".2 960.00 



115M5 1,271.51 115^2 1,260.00 



121°.25 1.540.36 12V A 1.560 00 



126°.24 1,796.98 125°.7 1,737.20 



130°. 33 2.030.33 129".3 1.936.90 



135°. 98 2.390.00 135°. 5 2,335.50 



138°.39 2,561.68 138'.0 2,538.00 



143°.07 2,923.53 U3M 2.937.60 



