greater as the temperature of distillation rises, yet the total production of tar is less. Mr. Wright's analysis,' showing the constituents of the various tars in the experiment just alluded to, is very instructive in this connection, for it shows that although the percentage of free carbon is augmented by increasing the temperature of distillation, a sufficient quantity of the lighter constituents of the tar is also decomposed into gas, to enable the increased proportion of carbon to be accounted for without falling back on the assumption that it is produced by the decomposition of illuminating hydrocarbons in the gas. With a make of about 7,000 cubic feet per ton, the tar contains some 9 per cent. of light naphtha, and a rather less proportion of light oil. When the make increases to about 10,000 cubic feet, the amount of these constituents falls to about one-third, or 3 per cent. of each respectively. At the highest temperature employed in Mr. Wright's experiments, not only is the quantity of light oil and naphtha reduced below. 1 per cent., but a large proportion of creosote oil is also decomposed. When the decrease in the total yield of tar is taken into consideration, it is evident that the gain in pitch is more than counterbalanced by loss of light oil, naphtha, and creosote.

Differences in the quality of the coal employed also influences the quantity and quality of the constituents found in the resulting tar. Newcastle coals produce tars rich in naphthalene and anthracene; Wigan coals yield tars containing much benzol and phenol. From the experiments of Bünte,' it would appear that the quantity of tar increases with the percentage of oxygen in the coal.

Tar derived from ordinary bituminous coal varies in specific gravity from 1·1, to 1-2, and the total yield of tar from coal may be said to be, on an average, 5 per cent. of 1 "Journ. Soc. Chem. Industry."

"Journal für Gasbeleuchtung," 1886.

the weight, the quantity produced per ton of Newcastle coals, at ordinary working temperatures, being about ten gallons, which is increased at low temperatures, as previously explained.

The composition of tar necessarily varies. The following is taken from Professor Wanklyn's "Gas Engineer's Chemical Manual."

The ultimate composition of a London tar (Professor Lewes) is approximately as follows:

With respect to the effects of temperature on the production of ammonia, Mr. Wright is in this case also the leading authority.

The following extracts give Mr. Wright's view on the subject (see Journal Soc. Chem. Industry," vol. vii., p. 59): "The results obtained from individual experiments, lasting twenty-four hours, have never been very determinate, and it appears that there is some cause, other than the temperature

of retorts, at present unknown, operating as a disturbing element.

"The averages of large numbers of experiments however always bring out the fact that at very low distillation temperatures the yield of ammonia is low, that a medium temperature brings out a maximum, and that beyond this, higher temperatures result in a slightly diminished production.

"The following case, where the coal (whose analysis has already been given) was employed, is typical of a great many that have been obtained, and will serve as well as a larger number to illustrate the tendency."

In this case the maximum yield of ammonia is with the make of 10,162 cubic feet per ton.

Mr. Wright further found that higher heats produced more bisulphide of carbon and sulphuretted hydrogen, but the product which was most remarkable with high heats was cyanogen, which was ten times as great with high heats as with low ones.

With regard to the production of coke, a high temperature generally produces a harder coke, but there are great differences in the character of the coke produced from the same coal, partly dependent on the temperature employed, and partly on the description of retort used.

F

Gas coke, although often spoken of as carbon, also contains nitrogen, sulphur, hydrogen, and oxygen.

Speaking generally, the products of distillation vary with the origin of the coal, Newcastle coals yielding a different class of tar to Derbyshire and South Yorkshire.

At the temperature usually employed in carbonization, the following are the products obtainable from a ton of Newcastle coal, as given by Professor Wanklyn in his admirable "Gas Engineer's Chemical Manual."

TH

CHAPTER VII.

PURIFICATION.

(Condensation.)

HE crude gas as it leaves the retorts may be said to enter the purification stage directly it reaches the hydraulic main, at which point the impurities consist of tarry bodies (impure hydrocarbons), carbonic acid, sulphuretted hydrogen, ammonia, cyanogen compounds, and sulphur compounds, other than sulphuretted hydrogen.

According to Professor Wanklyn,' gas, as it leaves the retorts," is accompanied by rather less than one-third of its weight of tar, by nearly half its weight of watery vapour, by about th of its weight of ammonia, by a little more thanth of its weight of carbonic acid, by from th to 2th of its weight of sulphuretted hydrogen, and by from 3th to th of its weight of sulphur in the condition of sulphuret of carbon and organo-sulphur compounds."

The first operation to which the gas is subjected is that of cooling, and as it cools, the vapours of various hydrocarbons and aqueous vapour, condense into the liquid form. The liquefied hydrocarbon vapours constitute what is known as tar, while the condensed aqueous vapour takes up ammonia, as well as CO2, SH2, and cyanogen, forming what is known as ammoniacal liquor. This cooling process commences immediately the gas reaches the retort mouthpiece, so that by the time it has arrived at the outlet

1 "The Gas Engineer's Chemical Manual.”