well known, were adopted for the purpose of obtaining 8 accurate results. When operated upon in this way, the only substances evolved from the salt were carbonic acid and water. As an instance of the results, 3.5 grains of the salt afforded 11.74 cubic inches of carbonic acid gas, and 0.9 of a grain of water. The mean of several experiments gave 32.93 cubic inches of carbonic acid gas, and 2.589 grains of water for every 10 grains of salt decomposed.
On these data, 100 grains of the salt would yield 329.3 cubic inches of carbonic acid, or 153.46 grain, equivalent to 41.9 grains of carbon, and 25.89 grains of water, equivalent to 2.877 grains of hydrogen. Hence 100 grains of the salt yielded—
| Baryta | 27.57 | 78 |
| Sulphuric acid | 30.17 | 85.35 |
| Carbon | 41.90 | 118.54 |
| Hydrogen | 2.877 | 8.13 |
| 102.517 |
In the second numerical column the experimental results are repeated, but increased, that baryta might be taken in the quantity representing one proportional, hydrogen being unity; and it will be seen that they do not differ far from the following theoretical statement:—
| Baryta | 1 | proportional | 78 |
| Sulphuric acid | 2 | ditto | 80 |
| Carbon | 20 | ditto | 120 |
| Hydrogen | 8 | ditto | 8 |
The quantity of sulphuric acid differs most importantly from the theoretical statement, and it probably is that element of the salt in the determination of which most errors are involved. The quantity of oxide of copper and of acids required to be used in that part of the analysis, may have introduced errors, affecting the small quantity of salt employed, which when multiplied, as in the deduction of the numbers above relative to 100 parts, may have created an error of that amount.
As there is no reason to suppose that during the combination of the acid with the baryta any change in its proportions takes place, the results above, minus the baryta, will represent its composition: from which it would appear, that one propor-
