OF ARTS AND SCIENCES. 25 



Reguault made ei"[ht observations on the maximum tension of sul- 

 phur vapor at temperatures varying from 387°.64 to 554°.03, and from 

 a discussion of these deduced the constants of an exponential formula, 

 by whicli he calculated a table of maximum tensions for every ten de- 

 grees between the extreme limits, and also plotted a cori-esponding 

 curve. It so happens, however, that the only two observations 

 within the range of ordinary atmospheric pressure fall outside, and 

 on the same side, of this assumed curve. These observations are the 

 ones usually taken as indicating the boiling point of sulphur ; and 

 Victor Meyer, in one of his methods of determining the density of the 

 vapors of substances wliich have a high boiling point, assumes a value 

 for the boiling point of sulphur (at the mean atmospheric pressure at 

 Zurich), which he obtains by simple interpolation from the two ob- 

 servations just referred to.* In like manner we have calculated the 

 above values corresponding to the pressures at which Mr. Bennett's 

 results were obtained on the basis of the same two observations ; but, 

 instead of simply interpolating by the first differences, we have as- 

 sumed that tlie variation between the two observed values would fol- 

 low the law indicated by the general curve, which Reguault gives as 

 the best expression for all his observations. But according as we take 

 the two observations, or the whole, we obtain values for the boiling 

 point of suljDhur differing by more than a degree ; and hence, as we 

 have already said, there is still an uncertainty in regard to the boiling 

 point to this extent. As is evident, Mr. Bennett's observations con- 

 firm very closely the interpretation of Regnault's results, adopted by 

 both Victor Meyer and by ourselves. 



After the accuracy of our method had been thus placed beyond 

 doubt within the limits required, Mr. Bennett made three determina- 

 tions of the boiling point of antimonious iodide, with the following re- 

 sults : — 



Barometer. Height at 0°, Boiliug Point of Sb T3. 



758.1 millimetres. A0O°A 



758.4 " 400°.9 



759.3 " 400°.9 



Probably only a small part of the difference between these observa 

 tions depends on the variations of pressure. We only regard the 

 method as accurate to whole degrees, and 401° is evidently the boiling 



* Fresenius's Zeitschrif t, xvi. 482. 



