The following tables are not arranged in the order in which the experiments were made. Table IV includes four experiments in which the current density was approximately 0.007 ampère per square centimetre of "active" cathode, or 1 ampère to 135 sq. cm. All the experiments gave the vacuum deposit higher than the air deposit, and the percentage differences agree fairly. The mean value = 0·16 per cent. at a mean temperature of 18°.1. Table V includes three experiments in which the current density was 0.006 ampère per square centimetre. It will be seen the vacuum deposit is higher than the air deposit in the three experiments. Mean value of difference 0.10 per cent. at a mean temperature of 16°.5. Table VI includes four experiments at a current density of 0.005 ampère per square centimetre. Again the vacuum deposit is higher than the corresponding air deposit. Mean value of difference = 0·14 per cent. at a mean temperature of 17°.5. Table VII includes five experiments at a current density of 0.004 ampère per square centimetre. All the vacuum deposits are higher than the corresponding air deposits. Mean value of the difference = 0·17 per cent. at a mean temperature of 19°. Table VIII includes five experiments at a current density of 0-0033 ampère per square centimetre. Again deposit in vacuo is higher than the deposit in air. Mean value of difference = 0·14 per cent. at a mean temperature of 15°.5. Table IX includes seven experiments at a current density of 0.0027 ampère per square centimetre. The same results obtained here as in the preceding tables; but the experiments of April 12 and 13 do not agree very well with the others. Mean value of difference = 0·16 per cent. at a mean temperature of 16°·6. Table X includes nine experiments in which the strength of current was approximately 0.01 ampère per square centimetre. This table is very interesting, as showing that near this particular current density the previously observed differences cease to exist. The nine experiments never gave a greater difference between corresponding deposits than 1 tenth-milligram. The voltameters were carefully tested for leakage before and after these experiments; and the experiment of April 7 (Table IX), in which there was a percentage difference of 0.12, follows that of April 5, without any alteration in the arrangement of the apparatus. In addition, the experiments of December 2 and 3 were made with a different bell-jar and a different arrangement from that used in the previous experiments. It may be fairly concluded that at this density no difference in the deposit exists. Table XI includes three experiments at a density of 0.02 ampère per square centimetre. It will be seen that there is practically no difference between the deposits. |