AND RATIO OK THKIR ATOMIC WKKIHTS. 7.'{ 



22. IITDKOGEN BY NKW METHOD. VERIFICATION OV ACCURACY OK WEIGHTS. 



The accuracy of the weights used for determining the. weights of hydrogen 

 received careful attention. I determined the rehition of each weight of two addi- 

 tional sets of small weig'hts to the kiloiiramme of the set marked K, on which ail 

 ray determinations of volume depend. Tlien determinations of weights of hydrogen 

 were made with all these four sets of weights ; so that it is improbable that any 

 serious eri-or is due to the uncertainty of the values of the weights employed. I 

 moreover determined the relation to the same kilogramme of two other sets of 

 small weights which were kept simply for the purpose of comparing them with the 

 weights which were used in weighings, so as to detect any accident to any of 

 these. 



. 23. HYDROGEN BY NEW METHOD. REDUCTION OF OBSERVATIONS. 



The readings of the barometer were corrected by means of the table given by 

 Landolt and Boernstein. The corrected numbers are called observed values. To 

 the capacities of the globes at 0° was added the capacity of that p)art of the con- 

 necting tubes which was always covered with ice. The volume contained by the 

 syphon barometer down to the mark noted was taken from the table on page 68, 

 and reduced to the volume which that amount of gas would have at 0°. The three 

 or four cubic centimetres which were not below the water level of the case sur- 

 rounding the barometer were assumed to be at the temperature of the barometer ; 

 which was near enough to the truth. The sum of these three volumes made up the 

 observed volume. The observed temperature is that of melting ice ; the elevation 

 of the cistern of the barometer introduces a correction of less than five thousandths 

 of a millimetre of mercury. We have therefore 



Weight of one litre of hydrogen _ W 760 G45 L 



at sea level, latitude 45° Y -j- v -\- v' P g 1 



24. HYDROGEN BY NEW METHOD. OBSERVATIONS AND RESULTS. 



The table gives the observations of the third and fourth sei'ies, together with 

 some observations which were unsuccessful on account of various slia-ht accidents 

 or unforeseen circumstances. Five columns define tlie volumes which are to be 

 added to the calibrated capacities of the globes ; these show the mark at which the 

 mercury stood in the barometei', the temperature here, the volume corresponding 

 this volume reduced to 0°, and the volume of the tubes which were covered with 

 ice. Next is shown the volume of the globes, the total volume, the pressure 

 reduced to 0°, but uncorrected for the length of my bar, the observed weight of 

 hydrogen, and the density computed for the sea level in latitude 45°. 



