182 PROCEEDINGS OP THE AMERICAN ACADEMY 



XII. 



CONTRIBUTIONS FROM THE CHEMICAL LABORATORY OF 

 HARVARD COLLEGE. — J. P. Cooke, Director. 



ADDITIONAL NOTE ON THE RELATIVE VALUES OF 



THE ATOMIC WEIGHTS OF HYDROGEN 



AND OXYGEN. 



By Josiah Parsons Cooke and Theodore William Richards. 



Presented March 14, 1888. 



The preceding paper on this subject was already in print, and a num- 

 ber of the extra copies had been distributed, when the writer received 

 a letter from Lord Rayleigh stating that he had been engaged on 

 a similar work, and had observed that the glass balloon used in Reg- 

 nault's method of weighing gas volumes, when exhausted, was sensibly 

 condensed by the pressure of the air. Obviously, if this were true, the 

 tare of the balloon thus exhausted would be too large in consequence 

 of the lessened buoyancy of the atmosphere, and hence the subse- 

 quently observed weight of gas when the balloon was filled would be 

 too small. A shrinkage amounting to a single cubic centimeter would 

 make a difference of about 1.29 milligrams, and Lord Rayleigh sug- 

 gested that our results might have been influenced by a constant error 

 arising from this source. As the same balloon represented in Figure 1 

 of the preceding paper had been used in all our determinations, and was 

 still in good condition, there was no difficulty in determining the amount 

 of shrinkage under exhaustion, and thus finding the correction which 

 ought to be applied to the results on this account. The method we 

 used was briefly as follows. 



The balloon was first exhausted, and then completely filled with 

 boiled distilled water at an observed temperature. The weight of this 

 water having been taken, and the internal volume of the balloon thus 

 determined, a small portion of the water — 190 cubic centimeters — 

 was run out, and the volume estimated both by direct measurement 

 and also by reweighing the balloon. With these data we could 

 readily calculate the volume of air left in the balloon for any given 

 temperature, and the small amount of water lost by evaporation in the 



