OF ARTS AND SCIENCES. 159 



weight can only be a very small fraction of the weight of the containing 

 vessel. Moreover, as the buoyancy of the air is fourteen and one half 

 times as great as the weight of the hydrogen, the variations in buoyancy 

 caused by changing atmospheric conditions have an all-important effect 

 on the apparent weight. The late Professor Regnault, of Paris, de- 

 vised, however, a very ingenious method of compensation, which could 

 readily be applied in this case. It consisted in balancing the globe 

 containing hydrogen, hung to one arm of the balance, by a second 

 globe of exactly the same external volume and made of the same mate- 

 rial, hung to the opposite arm ; and so arranging the balance case that 

 they should hang in the same enclosure, and therefore be equally 

 affected by atmospheric changes. This method was applied in the 

 problem before us ; and after a number of trials it was found possible 

 to make the compensation so accurate that under good conditions the 

 weight of a globe holding five litres of gas did not vary more than one 

 tenth of a milligram through large changes of temperature and pressure. 

 In order now to weigh hydrogen with this apparatus, it was only neces- 

 sary to exhaust the air from the glass receiver, and, after balancing it 

 as described, to fill it with pure gas, when the increased weight — less 

 than half a gram with our apparatus — was the weight of hydrogen 

 required. 



The balance employed was an excellent one, made about twenty 

 years ago by Becker and Sons, of New York. With a load of five 

 hundred grams in each pan, it turns very perceptibly with one tenth 

 of a milligram, and shows this small difference of weight with very 

 great constancy. 



The globe and its counterpoise were hung from hooks soldered to the 

 bottoms of the pans by means of wires which swung freely through small 

 holes made for the purpose through the bottom of the balance case, and 

 also through the top of the shelf on which the case stood. The enclosure 

 in which the globe and its counterpoise hung was a box made of tinned 

 iron fastened to the bottom of the shelf, and having doors in front like 

 an oven, through which the globe could be removed or hung in posi- 

 tion. This case was coated with lampblack on the inside, in order to 

 secure uniformity of temperature ; and the air was kept dry by means 

 of two large dishes of sulphuric acid, placed on shelves at the top of 

 the case. We first placed the sulphuric acid dishes on the bottom of 

 the case in the usual way, but we found it impossible thus to secure a 

 uniform condition of the atmosphere within ; and as moist air is neces- 

 sarily lighter than an equal volume of dry air at the same temperature 

 and pressure, it is obvious that any drying material will have the 



