ON THE ATOMIC WEIGHT OF CHLORINE. 195 



8. Results of the Experiments. 



In the following tables we have put together the results obtained in the nine 

 experiments. Table II. contains the volumes of the several residual gases, reduced 

 to normal temperature and pressure, as determined by the gas-analysis. 



We have assumed that the nitrogen found at the end of the experiments is due to 

 residual air left in the evacuation of the large combustion globe. It is conceivable 

 that a trace of this nitrogen came from the palladium bulb and was weighed as 

 hydrogen. If that were so, the atomic weight of chlorine we have found would be too 

 low. The total volume of nitrogen found in the nine experiments was '8 cub. centim. 

 To make an extreme supposition if all this nitrogen had been introduced from the 

 palladium bulb, and weighed as hydrogen, and therefore all the oxygen had come 

 from the steam, the atomic weight of chlorine found by us would be '005 too low. 



In Table III. we have put together the several portions of residual chlorine not 

 combining with the weighed hydrogen : (i.) that calculated from the iodine found in 

 the globe ; (ii.) that calculated from the iodine vapour drawn over with the residual 

 gases and caught in the wash-bottle; and (iii.) that calculated from the oxygen 

 found in the residual gases less the oxygen assumed (from the nitrogen) to be present 

 as air. 



In Table IV. the weights of the bulbs before and after the combustion are given, 

 with the corrections for buoyancy and for the unburnt gases. We set out the 

 hydrogen weighings to five places of decimals, although it is not, of course, suggested 

 that the absolute weight of the palladium bulb can be determined to this degree of 

 accuracy. The fifth figure does not affect the mean atomic weight deduced from the 

 experiments. 



2 C 2 



