or adjusted. Then a current was passed long enough to expel any air which might have entered, when it was again closed and weighed for the next experiment.

10.-DENSITY OF ELECTROLYTIC GAS.

EUDIOMETRIC ANALYSIS.

In the earlier experiments, a Toepler pump was connected to the globes as shown in Fig. 32. At a is the inclined tube leading to the globes, seen in its connection with them in Fig. 28, page 77. The U tube filled with mercury to shut off the pump is also shown. To this was connected the tube c, leading to the Toepler pump d. While the pressure

n

h

FIG. 32.-Toepler pump used to measure sample for analysis; small Toepler pump for transferring convenient quantity to eudiometer.

the pressure had been determined, the mercury was lowered till the bottom of the tube c was uncovered. Gas from the globes then passed through the tube to the Toepler pump, driving over the mercury in the tube c, but without doing damage. The pump body having been filled with gas, the volume withdrawn was determined 9 by a second reading of the pressure of the remaining gas. The mercury in the pump was then raised so as to separate this fraction from that left in the globes.

A Bunsen eudiometer was filled with mercury and set at m, Fig. 32. A small Toepler pump, having a capacity of, say, twenty cubic centimetres, had been connected as shown at 7, and was provided with a trap against the entrance of air with the mercury which actuated it. With this, the gas was transferred in suitable volumes to the eudiometer and exploded. Four explosions a minute was the usual working rate, so that the 1.6 litres which the pump held were exploded in forty or fifty minutes.

11.-DENSITY OF ELECTROLYTIC GAS. DIFFICULTY OF PREVENTING ENTRANCE OF AIR

INTO THE EUDIOMETER.

The eudiometer stood in mercury to the depth of about ten centimetres, but at first nitrogen was always found in the residue after explosion. Increased care did

not accomplish anything till the mercury in which the eudiometer stood was covered with a layer of sulphuric acid, after which no more nitrogen was found. It was therefore fair to assume that all the nitrogen found had been due to the entrance of air at the last operation; the volume of oxygen accompanying it was therefore computed and double this volume of hydrogen was assumed to have disappeared by combination with this oxygen derived from the air.

12.-DENSITY OF ELECTROLYTIC GAS.

LEAKAGE OF AIR BETWEEN MERCURY AND

GLASS.

It may be said that the contact of mercury with glass is proof against leakage only when it is nearly as perfect as in a barometric tube. A joint which leaks a little may not be perceptibly improved by immersion in mercury. I have seen a joint between a rubber and a glass tube, which leaked into a vacuum at the rate of a bubble a minute, keep on at apparently the same rate when immersed ten centimetres in mercury, because the contact between the glass and the mercury was not sufficiently perfect. Since a clean surface is hard to maintain on the outside of a tube, the only way in which mercury can be safely assumed really to stop a leak is, by using it to carry a layer of sulphuric acid, as was indicated by Crookes in his experiments with high vacua.

13.-DENSITY OF ELECTROLYTIC GAS.

OXIDATION OF MERCURY. LARGE EUDIOMETER.

In some of the eudiometric analyses, some oxidation of mercury was noticed, as might be expected. The amount of oxygen thus consumed was obviously neg ligible, but it was thought well to avoid possible objection by making it less. A measured volume of hydrogen was therefore introduced into the eudiometer before the admission of the electrolytic mixture. The results thus obtained were the same as before within the errors of observation.

I then constructed a large eudiometer in which all but a hundredth or a thousandth of the gas taken for analysis could be exploded out of contact with mercury. It has a capacity of 3.2 litres. A stopper ground into its place and extending down through ninety centimetres of mercury is so arranged that it can be raised or lowered at will, without danger of leakage. When an explosion was to be made, the passage was stopped by this contrivance, so that the explosion did not reach to the mercury. This eudiometer will safely bear the explosion of eight hundred cubic centimetres of the mixed gases as measured at atmospheric

pressure.

The admission of gas to the eudiometer was controlled by the stopper b, Fig. 30. When the desired pressure was reached, b was closed and then the similar

stopper of the large eudiometer, after which the spark was passed. The two stoppers were then again opened, and the process repeated from four to twenty times. At the end of the process the pressure in the globes was observed, thus determining the volume withdrawn for analysis. The globes were kept surrounded with ice during the whole course of the experiments.

It is obvious that the measurement and analysis of the residue could not be effected in the large eudiometer. This was therefore connected by a leakage-proof mercurial valve, with a self-acting Toepler pump. When the last explosion had been made, this valve was opened, and the pump set in action. The large quantity of water formed during the explosion had to be absorbed between the eudiometer and the pump. A special form of drying tube was constructed, but it is hardly worth while to describe it. It may suffice to say that the drying agent was pure sulphuric acid boiled in a vacuum till it did not give off a harmful quantity of absorbed gas, and that this acid could be flowed over the glass beads filling the drying chamber. Even after absorbing fifty grammes of water, it was so effective

that a vacuum of two or three millionths could be attained.

From the Toepler pump, the gas was transferred to a Bunsen eudiometer. Owing to the fact that the explosion in the large eudiometer did not extend to the surface of the mercury, there was always enough uncombined oxygen and hydrogen in the residue to admit of satisfactory explosion, after which the residue was analyzed.

14.-DENSITY OF MIXED GASES.

CAN ELECTROLYTIC GAS BE OBTAINED IN ATOMIC

PROPORTIONS?

In all my experiments, I found an excess of hydrogen. This is due to secondary reactions in the electrolytic cell, attended with consumption of oxygen. Now it is one of the innumerable possible cases that there should be an equilibrium between the production and destruction of the oxide formed, so that for a time the two gases should be delivered in atomic proportions. But all the precautions which occurred to me were used in vain. The voltameter was treated in a manner perfectly uniform, and used only at the temperature of melting ice, but there was always an excess of hydrogen.

Once, to learn a little about the stability of any supposed equilibrium, I connected the voltameter directly to the delivery tube leading to a Bunsen eudiometer containing a measured quantity of hydrogen, designed to prevent oxidation of mercury. The eudiometer stood deeply in mercury covered with sulphuric acid. Electrolytic gas was introduced and exploded for many times; the amount of hydrogen gradually increased. The current was now increased so as to warm the

eudiometer about twenty degrees. The hydrogen in the eudiometer soon began to decrease as the explosions were repeated, showing that now oxygen was in excess.

The equilibrium was so easily disturbed that some doubt cannot but be felt. whether Leduc obtained the equilibrium which he hoped for.

15.-DENSITY OF ELECTROLYTIC GAS. OBSERVATIONS ON THE EXCESS OF HYDROGEN. The determinations of the excess of hydrogen above the atomic ratio in the mixtures whose density was determined, are given in the following table. In the first four experiments, the Bunsen eudiometer was used; in the others, the large eudiometer containing 3.2 litres. The first column gives the pressure originally noted; the second, the diminution of pressure caused by taking out a part for analysis, the third the excess of hydrogen found in this part; and the last, the ratio of this hydrogen to the whole volume analyzed.

Experiment.

Pressure.

Diminution of

pressure.

3.05

.00033
.00036

.0002 I

.00022

.000 29

.00037

.00042

The mean is .000293. This is the ratio of the excess of hydrogen to the whole combined volumes of oxygen and of hydrogen. Multiplying by the factor 3, we have .00088, a correction to be applied to the ratio of the hydrogen to the oxygen of the mixture in order to obtain the ratio of the volumes of hydrogen and oxygen which would combine without residue.

16.-DENSITY OF ELECTROLYTIC GAS. OBSERVATIONS.

The following table gives the particulars of the determinations of the density of the mixture of oxygen and hydrogen obtained by the electrolysis of sodium hydroxide. Seven columns give the mark at which the mercury stood in the barometer, the temperature here, the volume corresponding, this volume reduced to 0°, the volume of the ice-covered connecting tubes, the volume of the small connecting tube near the voltameter, and the volumes of the globes. Following these are the columns which give the total volume, the pressure, the weight of the mixed gases, and the density computed for the sea level in latitude 45°, according to the formula

I 212.8

20.9° 65.5

60.8

41.1

3.1

2

211.0

19.0

66.0

61.7

41.1

34

3

2109

19.5

43.2587 43.3637 746.16 22.7858 .535441 43.2587 43-3649 754.90 23.0541 .535459 43.2587 43.3646 748.93 22.8770 .535584 43.2587 43.3644 753.20 23.0027 -535477 43.2587 43.3640 748.20 22.8521 .535531 43.2587 43.3645 752.15 22.9723 535514 43.2587 43.3639 753.36 23.0064 .535456 43.2587 43-3557 772.17 23.5798 .535434 43.2587 43.3638 753.39 23.0098 .535515 43.2587 43.3628 753.97 23.0269 .535513

If we increase the mean by one thirty-thousandth,* we have

D = 0.535510 gr. ± 0.000010.

17.-DENSITY OF ELECTROLYTIC GAS. REDUCTION OF RESULTS,

1 2+

If, now, Boyle's law applied to this mixture of gases, we should have, putting H, O, M for the three densities involved, and 4x for the ratio of the volumes of oxygen and hydrogen which combine, measured at standard temperature and pressure,

If we may judge by the brief account of Leduc's experiments given in the Comptes Rendus, this reduction was adopted by him. But the reduction takes no account of the deviations of the gases from Boyle's law. We require to know what would be the pressure of one volume of oxygen and two volumes of hydrogen when mixed and made to occupy three volumes. Until direct observations are made, we must compute from observations of the volume and pressure of the separate gases at different pressures. Since such observations are represented by the equation of Van der Waals with little precision, we may only hope that a reduction founded on it may temporarily serve as an approximation. Writing capitals for oxygen and small letters for hydrogen, we have

Putting the original volume and pressure equal to unity, we get